INDEX

Explanations

governmental department names and associated entities

oai_token-act-pair · gpt-3.5-turbo

instances of the word "and" 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

rises

-0.77

aunders

-0.76

helps

-0.75

achev

-0.72

>[

-0.71

hello

-0.71

gart

-0.69

ugi

-0.68

keyes

-0.68

:[

-0.68

POSITIVE LOGITS

 Associates

0.94

 Environment

0.88

 Accountability

0.84

 Related

0.83

 Sciences

0.81

 Resources

0.80

 Other

0.80

 Development

0.79

 Environmental

0.78

 Telecommunications

0.77

Activations Density 0.174%

governmental department names and associated entities

instances of the word "and" in various contexts

No Comments

No Known Activations