INDEX

Explanations

proper nouns related to people and places

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

instances of the term "ne" 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 @ 12-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.hook_resid_post

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.11.hook_resid_post

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ENCY

-0.67

ERSON

-0.67

 retard

-0.67

DOWN

-0.59

hips

-0.59

 pains

-0.59

 lengths

-0.58

 tempered

-0.57

 complete

-0.56

 hoop

-0.56

POSITIVE LOGITS

uve

1.07

volent

0.99

arthed

0.97

cker

0.94

arest

0.93

uron

0.93

farious

0.92

umatic

0.91

kens

0.91

cks

0.91

Activations Density 0.041%

proper nouns related to people and places

instances of the term "ne" in various contexts

No Comments

No Known Activations