INDEX

Explanations

instances where articles and posts are referenced or introduced

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

words related to articles and written content

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

unt

-0.68

phony

-0.67

ear

-0.66

ience

-0.65

 Others

-0.65

dayName

-0.62

inel

-0.61

ivid

-0.61

imental

-0.61

soType

-0.61

POSITIVE LOGITS

 however

0.78

 titled

0.76

 Carlo

0.69

 readers

0.68

hov

0.68

 moreover

0.66

KB

0.64

 Junk

0.63

we

0.62

 subscript

0.61

Activations Density 0.227%

instances where articles and posts are referenced or introduced

words related to articles and written content

No Comments

No Known Activations