INDEX

Explanations

words that indicate doubt or uncertainty

oai_token-act-pair · gpt-3.5-turbo

negative contractions indicating unfulfilled actions or states

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

CVE

-0.69

Reviewer

-0.63

 behavi

-0.61

è»

-0.60

 SetTextColor

-0.60

 Polaris

-0.58

CRIP

-0.58

å£«

-0.58

 Pike

-0.57

çĦ

-0.57

POSITIVE LOGITS

aken

1.00

alion

0.99

ween

0.97

reprene

0.95

ournament

0.94

akers

0.92

rees

0.91

otally

0.89

itles

0.88

enegger

0.87

Activations Density 0.013%

words that indicate doubt or uncertainty

negative contractions indicating unfulfilled actions or states

No Comments

No Known Activations