INDEX

Explanations

mentions of and actions related to the term "fork"

oai_token-act-pair · gpt-3.5-turbo

references to various types of "forks" in different contexts

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

useum

-0.72

 Penet

-0.69

roit

-0.66

 Scotia

-0.64

pson

-0.64

berman

-0.64

umbai

-0.61

cester

-0.61

yre

-0.60

ACP

-0.60

POSITIVE LOGITS

lift

1.46

edIn

0.90

 fork

0.90

ingen

0.82

nect

0.81

ners

0.78

 forks

0.76

Join

0.75

pit

0.74

fork

0.73

Activations Density 0.010%

mentions of and actions related to the term "fork"

references to various types of "forks" in different contexts

No Comments

No Known Activations