INDEX

Explanations

words related to geographical locations, particularly 'shore.'

oai_token-act-pair · gpt-3.5-turbo

references to offshore environments and related terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

 authentic

-0.64

imates

-0.64

 prohibition

-0.64

 contiguous

-0.63

 abstinence

-0.62

icate

-0.59

 ours

-0.58

 buckle

-0.57

 extinct

-0.57

 Franks

-0.57

POSITIVE LOGITS

roleum

0.90

shore

0.85

enment

0.80

abases

0.80

lich

0.79

xual

0.78

akespeare

0.78

å£«

0.77

tenance

0.77

man

0.76

Activations Density 0.016%

words related to geographical locations, particularly 'shore.'

references to offshore environments and related terms

No Comments

No Known Activations