INDEX

Explanations

phrases introducing or summarizing information

oai_token-act-pair · gpt-3.5-turbo

statements that introduce clarifications or explanations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

"},"

-0.68

 greater

-0.64

 Kard

-0.62

ger

-0.60

 Frie

-0.59

pent

-0.59

 visitation

-0.57

 imminent

-0.56

Bac

-0.55

lor

-0.55

POSITIVE LOGITS

 nutshell

1.03

etheless

0.96

aredevil

0.82

Basically

0.80

stream

0.79

 Basically

0.78

ovych

0.77

atel

0.77

creen

0.75

 Essentially

0.74

Activations Density 0.008%

phrases introducing or summarizing information

statements that introduce clarifications or explanations

No Comments

No Known Activations