INDEX

Explanations

words related to strong negative emotions or reactions

oai_token-act-pair · gpt-3.5-turbo

expressions of anger or frustration

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

acea

-0.88

erva

-0.85

cius

-0.82

oak

-0.81

arette

-0.78

querque

-0.77

win

-0.77

elia

-0.76

pty

-0.75

ynski

-0.75

POSITIVE LOGITS

idious

0.74

 ultras

0.70

 Furious

0.68

 Attacks

0.64

Dug

0.64

ãĥ£

0.62

quished

0.61

icago

0.61

 furious

0.61

err

0.60

Activations Density 0.040%

words related to strong negative emotions or reactions

expressions of anger or frustration

No Comments

No Known Activations