INDEX

Explanations

"Submitted by" statements at the end of documents

oai_token-act-pair · gpt-3.5-turbo

the word "Sub" indicating submissions, likely related to media content or contributions

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

Dol

-0.79

INESS

-0.77

 Cors

-0.76

 Papers

-0.74

 Brooks

-0.74

ILLE

-0.73

 diplom

-0.72

fir

-0.70

rot

-0.70

 Zurich

-0.69

POSITIVE LOGITS

sequent

1.55

sequently

1.51

mitted

1.47

stantial

1.40

ordinate

1.30

mitting

1.21

verting

1.20

stant

1.15

jug

1.15

scription

1.14

Activations Density 0.017%

"Submitted by" statements at the end of documents

the word "Sub" indicating submissions, likely related to media content or contributions

No Comments

No Known Activations