INDEX

Explanations

introductory phrases indicating the beginning of a new topic or statement

oai_token-act-pair · gpt-3.5-turbo

introductory phrases that signal the start of a list or explanation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ween

-0.85

kids

-0.76

sung

-0.74

mens

-0.71

abled

-0.71

lain

-0.71

sports

-0.70

anim

-0.69

thirst

-0.69

breeding

-0.68

POSITIVE LOGITS

 introdu

0.89

 volley

0.89

 foremost

0.76

 admit

0.72

 congratulations

0.71

 apologies

0.69

ega

0.68

 apologize

0.67

 lesson

0.65

 premise

0.65

Activations Density 0.115%

introductory phrases indicating the beginning of a new topic or statement

introductory phrases that signal the start of a list or explanation

No Comments

No Known Activations