INDEX

Explanations

dates

oai_token-act-pair · gpt-3.5-turbo

negations or phrases indicating the absence of something

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Reviewer

-0.82

 behavi

-0.69

estern

-0.68

 Reloaded

-0.66

soType

-0.65

CVE

-0.65

 Sparrow

-0.64

ħĭ

-0.64

è»

-0.63

 Penguin

-0.62

POSITIVE LOGITS

cha

1.10

 necessarily

0.94

urtle

0.92

otally

0.92

ional

0.91

itles

0.90

ople

0.90

acular

0.89

unes

0.88

ween

0.87

Activations Density 0.107%

dates

negations or phrases indicating the absence of something

No Comments

No Known Activations