INDEX

Explanations

references to specific locations and positions

oai_token-act-pair · gpt-3.5-turbo

proper nouns and significant entities within a context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

ibilities

-0.70

pmwiki

-0.63

=/

-0.61

Pak

-0.58

rawdownloadcloneembedreportprint

-0.58

lihood

-0.54

./

-0.53

ivating

-0.52

cipled

-0.51

lessly

-0.51

POSITIVE LOGITS

ensis

0.61

senal

0.59

 juven

0.59

 Bows

0.57

Osw

0.55

iatus

0.55

 Bowen

0.55

Khe

0.54

Pax

0.54

Sao

0.54

Activations Density 1.272%

references to specific locations and positions

proper nouns and significant entities within a context

No Comments

No Known Activations