INDEX

Explanations

informational phrases indicating where to find more details or sources on a specific topic

oai_token-act-pair · gpt-3.5-turbo

references to additional information or resources

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

 ãĤµãĥ¼ãĥĨãĤ£ãĥ¯ãĥ³

-0.71

ĳå£«

-0.69

 Takeru

-0.69

 plurality

-0.67

wagen

-0.62

 Yugoslavia

-0.61

 nostalg

-0.60

 majority

-0.59

 blender

-0.59

 majorities

-0.58

POSITIVE LOGITS

 related

0.79

than

0.74

related

0.74

amazon

0.73

natureconservancy

0.73

 Helpful

0.71

usat

0.71

ocations

0.71

 than

0.70

www

0.70

Activations Density 0.149%

informational phrases indicating where to find more details or sources on a specific topic

references to additional information or resources

No Comments

No Known Activations