INDEX

Explanations

different types or categories of objects or concepts

oai_token-act-pair · gpt-3.5-turbo

mentions of diversity and different categories or classifications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

arton

-0.64

mobi

-0.61

jew

-0.61

efficiency

-0.61

Brave

-0.58

iron

-0.56

coal

-0.56

ayn

-0.56

 Adamant

-0.56

requisite

-0.55

POSITIVE LOGITS

 differing

1.00

 depending

0.99

 different

0.96

 imaginable

0.94

paces

0.93

 configurations

0.91

 sexes

0.85

 vying

0.84

 varying

0.84

 styles

0.83

Activations Density 0.249%

different types or categories of objects or concepts

mentions of diversity and different categories or classifications

No Comments

No Known Activations