INDEX

Explanations

proper nouns or names related to political figures

oai_token-act-pair · gpt-3.5-turbo

occurrences of the letter 'h'

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre

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

 Majority

-0.62

 incorpor

-0.59

 proposition

-0.58

 Bates

-0.57

 Proposition

-0.56

ificate

-0.56

 seminal

-0.55

 judged

-0.55

 establishment

-0.54

 vacancy

-0.53

POSITIVE LOGITS

3.78

hin

2.17

hm

2.04

hs

1.97

hr

1.96

hn

1.94

hw

1.94

hu

1.93

hur

1.90

hang

1.90

Activations Density 0.026%

proper nouns or names related to political figures

occurrences of the letter 'h'

No Comments

No Known Activations