INDEX

Explanations

instances related to speech or communication, specifically focusing on the word 'Speakers'

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "Speak" in various forms and 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

etheless

-0.89

MpServer

-0.74

 guiActiveUnfocused

-0.72

 Leilan

-0.72

sym

-0.71

enegger

-0.70

imaru

-0.70

maxwell

-0.69

ORED

-0.68

hap

-0.68

POSITIVE LOGITS

Spe

1.05

eches

0.99

spe

0.95

lling

0.94

Spe

0.90

edy

0.88

eding

0.85

culation

0.85

akers

0.83

ciation

0.83

Activations Density 0.010%

instances related to speech or communication, specifically focusing on the word 'Speakers'

occurrences of the word "Speak" in various forms and contexts

No Comments

No Known Activations