INDEX

Explanations

phrases related to progress, achievement, and action

oai_token-act-pair · gpt-3.5-turbo

references to processes and actions related to progress and achievement

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

Downloadha

-0.63

 herself

-0.59

 clipboard

-0.59

 Digest

-0.58

Usage

-0.57

CRC

-0.56

Pain

-0.55

oshenko

-0.54

tek

-0.54

 himself

-0.54

POSITIVE LOGITS

 ourselves

1.57

our

0.88

 ours

0.78

OUR

0.73

 selves

0.72

 asses

0.71

mble

0.70

assies

0.69

yg

0.68

 together

0.66

Activations Density 0.911%

phrases related to progress, achievement, and action

references to processes and actions related to progress and achievement

No Comments

No Known Activations