INDEX

Explanations

negative sentiments or criticism

oai_token-act-pair · gpt-3.5-turbo

phrases that describe a negative or undesirable quality

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

ourke

-0.81

©¶æ¥µ

-0.68

¥

-0.65

irrel

-0.62

ļéĨĴ

-0.60

âĹ¼

-0.60

 Pione

-0.60

SHIP

-0.59

 Citation

-0.59

 Million

-0.58

POSITIVE LOGITS

uminati

1.52

ogical

1.48

iberal

1.40

umin

1.34

usive

1.31

iquid

1.29

inois

1.15

usional

1.04

awar

1.00

icit

0.94

Activations Density 0.014%

negative sentiments or criticism

phrases that describe a negative or undesirable quality

No Comments

No Known Activations