INDEX

Explanations

code snippets and programming terms

oai_token-act-pair · gpt-3.5-turbo

sequences of repeated letters or patterns

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 ÂŃ

-0.81

 MILL

-0.75

 Fallout

-0.75

 Patriots

-0.74

 BART

-0.73

 Deadpool

-0.73

 HuffPost

-0.72

 Avengers

-0.72

 Craigslist

-0.72

 Consumer

-0.71

POSITIVE LOGITS

tein

1.35

tk

1.25

kas

1.25

nown

1.21

scl

1.21

inav

1.20

mr

1.19

vae

1.19

aq

1.18

oqu

1.18

Activations Density 0.307%

code snippets and programming terms

sequences of repeated letters or patterns

No Comments

No Known Activations