INDEX

Explanations

words related to formal or official actions taken towards a particular cause or issue

oai_token-act-pair · gpt-3.5-turbo

words that indicate transformation or conversion processes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

 boun

-0.72

 subur

-0.68

 fare

-0.68

 peril

-0.68

omnia

-0.65

sle

-0.65

srf

-0.64

 exha

-0.63

affer

-0.63

 carbohyd

-0.63

POSITIVE LOGITS

ments

1.20

ings

0.96

ment

0.95

ABLE

0.95

ously

0.91

INGS

0.89

ify

0.88

LECT

0.86

ISE

0.85

able

0.84

Activations Density 0.101%

words related to formal or official actions taken towards a particular cause or issue

words that indicate transformation or conversion processes

No Comments

No Known Activations