INDEX

Explanations

strong adjectives or descriptive words

oai_token-act-pair · gpt-3.5-turbo

terms associated with specific categories or classifications

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

Thumbnail

-0.60

leon

-0.58

allery

-0.56

lished

-0.56

aminer

-0.54

hyde

-0.52

Ö¼

-0.52

 âĶľ

-0.52

uana

-0.51

Ern

-0.49

POSITIVE LOGITS

 buffs

0.58

pes

0.55

 immunity

0.53

isively

0.53

bugs

0.52

coins

0.52

 probes

0.52

ickets

0.49

aram

0.49

ans

0.49

Activations Density 0.964%

strong adjectives or descriptive words

terms associated with specific categories or classifications

No Comments

No Known Activations