INDEX

Explanations

words related to strange or unconventional characteristics

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "weird" and its variants

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

vation

-0.91

ptive

-0.89

Interstitial

-0.89

ailable

-0.84

ptives

-0.83

adr

-0.82

cussion

-0.79

aders

-0.78

FORE

-0.77

idation

-0.77

POSITIVE LOGITS

ness

1.15

ly

1.08

nesses

0.95

ety

0.90

entimes

0.84

est

0.81

oes

0.76

balls

0.75

 Weird

0.75

 weird

0.74

Activations Density 0.010%

words related to strange or unconventional characteristics

occurrences of the word "weird" and its variants

No Comments

No Known Activations