INDEX

Explanations

mentions of technology-related terms, specifically related to machines and their functionalities

oai_token-act-pair · gpt-3.5-turbo

references to computers and machines

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

alez

-0.89

acious

-0.83

umper

-0.81

iosyncr

-0.77

ibilities

-0.76

raints

-0.73

ript

-0.71

uds

-0.70

uating

-0.68

aciously

-0.68

POSITIVE LOGITS

eers

0.80

gun

0.79

 machines

0.76

 readable

0.73

guns

0.72

eering

0.71

 reproduce

0.71

chnology

0.70

imbabwe

0.70

 Takeru

0.69

Activations Density 0.012%

mentions of technology-related terms, specifically related to machines and their functionalities

references to computers and machines

No Comments

No Known Activations