INDEX

Explanations

the word "himself" followed by a proper noun, likely related to individuals taking specific actions or being involved in certain contexts

oai_token-act-pair · gpt-3.5-turbo

the word "himself" and its emphasis in various contexts

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.3.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ammy

-0.69

 Sierra

-0.68

 Syndicate

-0.68

onal

-0.68

sweet

-0.67

grade

-0.66

grain

-0.65

emis

-0.65

heny

-0.64

cemic

-0.63

POSITIVE LOGITS

 tremend

0.75

selves

0.74

 ashamed

0.74

 submar

0.73

 profess

0.69

åĤ

0.68

 underwater

0.67

 worshipped

0.67

 guarded

0.67

 creatively

0.66

Activations Density 0.060%

the word "himself" followed by a proper noun, likely related to individuals taking specific actions or being involved in certain contexts

the word "himself" and its emphasis in various contexts

No Comments

No Known Activations