INDEX

Explanations

words related to importance or severity

oai_token-act-pair · gpt-3.5-turbo

intensifiers or descriptors indicating significant issues or threats

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

 Leilan

-0.71

 Reef

-0.66

 Kirin

-0.61

Ley

-0.61

 Cath

-0.60

Kra

-0.60

 Shepard

-0.60

Aus

-0.60

 Weston

-0.59

 Muss

-0.59

POSITIVE LOGITS

theless

1.13

terday

1.03

usterity

1.03

tenance

0.97

mosp

0.91

etheless

0.89

gettable

0.88

veyard

0.87

selves

0.86

withstanding

0.86

Activations Density 0.220%

words related to importance or severity

intensifiers or descriptors indicating significant issues or threats

No Known Activations