INDEX

Explanations

terms related to expertise, particularly in a specific field or hobby

oai_token-act-pair · gpt-3.5-turbo

suffixes related to adjectives and nouns, particularly focusing on the themes of identity and categorization

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

melon

-0.76

gression

-0.75

stable

-0.70

agan

-0.66

jack

-0.65

powers

-0.65

mits

-0.64

 gestation

-0.64

 ILCS

-0.62

ãģĨ

-0.62

POSITIVE LOGITS

naire

0.81

ures

0.76

antes

0.75

alist

0.73

 Bonds

0.72

 Garc

0.72

eral

0.71

ielding

0.69

redo

0.69

arette

0.69

Activations Density 0.026%

terms related to expertise, particularly in a specific field or hobby

suffixes related to adjectives and nouns, particularly focusing on the themes of identity and categorization

No Comments

No Known Activations