INDEX

Explanations

appearances of structured text patterns used in programming or data-related contexts

oai_token-act-pair · gpt-3.5-turbo

special character sequences or symbols that are not part of standard text

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 bount

-0.52

 capsule

-0.51

è£ħ

-0.48

 microbiome

-0.48

 offshore

-0.47

 inform

-0.46

 estim

-0.46

 pres

-0.46

usalem

-0.45

 similar

-0.45

POSITIVE LOGITS

-|

0.70

ONSORED

0.68

 Ibid

0.64

Too

0.62

Coun

0.62

Unless

0.61

Certainly

0.61

Again

0.60

He

0.60

Sure

0.58

Activations Density 0.515%

appearances of structured text patterns used in programming or data-related contexts

special character sequences or symbols that are not part of standard text

No Comments

No Known Activations