INDEX

Explanations

data related to ratings and classifications

oai_token-act-pair · gpt-3.5-turbo

references to classification codes related to different entities or items

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

 Panel

-0.59

 Pitch

-0.58

 Morales

-0.58

 pitcher

-0.57

Pak

-0.57

uez

-0.57

RIS

-0.56

 Rapids

-0.55

 stereotyp

-0.55

 Ribbon

-0.55

POSITIVE LOGITS

1.24

1.21

1.20

1.17

1.14

1.12

1.10

1.09

Activations Density 0.062%

data related to ratings and classifications

references to classification codes related to different entities or items

No Comments

No Known Activations