INDEX

Explanations

detailed descriptions or information

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "description" and related terms in various contexts

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

nuts

-0.76

ced

-0.75

yrus

-0.70

oner

-0.70

 scrimmage

-0.69

avis

-0.63

estern

-0.63

vernment

-0.63

riel

-0.62

abb

-0.62

POSITIVE LOGITS

 description

0.88

 thereof

0.85

 descriptions

0.82

description

0.79

hyde

0.77

 synopsis

0.76

Description

0.75

 wording

0.74

cloth

0.73

anguage

0.73

Activations Density 0.028%

detailed descriptions or information

occurrences of the word "description" and related terms in various contexts

No Comments

No Known Activations