INDEX

Explanations

phrases related to public speaking or being quoted in interviews

oai_token-act-pair · gpt-3.5-turbo

instances of the word "Speaking" 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 @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

-0.78

retty

-0.72

ascade

-0.70

======

-0.69

ifle

-0.68

=~

-0.68

adr

-0.67

\/\/

-0.67

aden

-0.67

Ã¨re

-0.66

POSITIVE LOGITS

 aloud

0.97

Speaking

0.82

 Speaking

0.82

 Loud

0.80

 volumes

0.80

voice

0.79

 spoke

0.72

 fluent

0.71

ILA

0.71

 engagements

0.70

Activations Density 0.009%

phrases related to public speaking or being quoted in interviews

instances of the word "Speaking" in various contexts

No Comments

No Known Activations