INDEX

Explanations

names containing the letter combination "ele"

oai_token-act-pair · gpt-3.5-turbo

references to elephants

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

DPR

-0.87

 Kers

-0.75

 Kenobi

-0.73

Present

-0.71

DERR

-0.69

Adv

-0.68

displayText

-0.66

EED

-0.66

 Papers

-0.66

Fuk

-0.66

POSITIVE LOGITS

ele

1.29

phant

1.08

venth

1.00

ghan

0.93

izabeth

0.89

fter

0.85

ven

0.85

ttes

0.81

bies

0.79

theless

0.78

Activations Density 0.008%

names containing the letter combination "ele"

references to elephants

No Comments

No Known Activations