INDEX

Explanations

studies or research papers published in academic journals

oai_token-act-pair · gpt-3.5-turbo

references to academic publications and proceedings

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

 lockout

-0.64

ghan

-0.62

 tears

-0.62

glers

-0.61

andi

-0.60

elli

-0.60

haar

-0.59

 heel

-0.57

uca

-0.57

hower

-0.57

POSITIVE LOGITS

 Proceedings

0.90

 Biology

0.79

ysical

0.79

 IEEE

0.79

 Academy

0.74

 Transactions

0.73

 Interface

0.73

ceed

0.72

 Biological

0.70

 congress

0.70

Activations Density 0.077%

studies or research papers published in academic journals

references to academic publications and proceedings

No Comments

No Known Activations