INDEX

Explanations

terms related to automotive maintenance and repair

oai_token-act-pair · gpt-3.5-turbo

references to treatment and medical conditions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

word

-0.65

watching

-0.63

arta

-0.59

intuitive

-0.56

oft

-0.56

ï

-0.54

FILE

-0.54

Whether

-0.53

Hour

-0.53

Ó

-0.53

POSITIVE LOGITS

 remainder

1.17

 rest

1.10

 remaining

1.03

 other

1.01

 others

0.98

 neighbouring

0.94

 adjoining

0.87

 OTHER

0.86

 Others

0.86

 larger

0.82

Activations Density 0.779%

terms related to automotive maintenance and repair

references to treatment and medical conditions

No Comments

No Known Activations