INDEX

Explanations

the word "No" with varying activation levels

oai_token-act-pair · gpt-3.5-turbo

Occurrences of the term "No" 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 @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

staking

-0.69

IENT

-0.65

realDonaldTrump

-0.65

 priceless

-0.64

lycer

-0.63

ULAR

-0.59

RAFT

-0.59

theless

-0.57

iership

-0.57

rolled

-0.57

POSITIVE LOGITS

isy

1.14

xious

1.12

zzle

1.06

vel

1.02

emi

1.02

onday

1.01

AH

0.92

ct

0.92

omi

0.92

isec

0.92

Activations Density 0.039%

the word "No" with varying activation levels

Occurrences of the term "No" in various contexts

No Comments

No Known Activations