INDEX

Explanations

phrases expressing opinions or judgements

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate making a statement or providing examples

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

xual

-0.63

iru

-0.63

uzz

-0.62

 Completed

-0.62

elsen

-0.61

arnaev

-0.60

fram

-0.60

 sidx

-0.58

resa

-0.58

enegger

-0.57

POSITIVE LOGITS

laughs

0.73

 nothing

0.71

 least

0.67

hem

0.62

Credit

0.57

_>

0.57

pecially

0.56

 detract

0.56

 suffice

0.56

gap

0.55

Activations Density 0.087%

phrases expressing opinions or judgements

phrases that indicate making a statement or providing examples

No Comments

No Known Activations