INDEX

Explanations

measurements or actions related to measuring physical attributes

oai_token-act-pair · gpt-3.5-turbo

terms related to measurement and quantification

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

illard

-0.86

ovie

-0.76

ategor

-0.75

ccoli

-0.73

lishing

-0.73

udeb

-0.71

jet

-0.71

redo

-0.70

chrome

-0.69

burning

-0.68

POSITIVE LOGITS

Measure

0.95

 Measure

0.89

 measuring

0.82

 measurement

0.80

 glim

0.79

 measurements

0.79

 measure

0.76

 Telescope

0.72

 instruments

0.72

 metrics

0.72

Activations Density 0.030%

measurements or actions related to measuring physical attributes

terms related to measurement and quantification

No Comments

No Known Activations