INDEX

Explanations

negative sentiments or contrasts within a given context

oai_token-act-pair · gpt-3.5-turbo

negations or expressions of absence

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

©¶æ

-0.80

ħĭ

-0.78

£ı

-0.71

Ĥİ

-0.71

 quieter

-0.64

 Moonlight

-0.63

Ń·

-0.61

Ļ½

-0.61

¿½

-0.61

 Carnage

-0.60

POSITIVE LOGITS

existent

1.14

hin

1.01

.","

0.90

 bother

0.89

 exist

0.89

 altogether

0.81

epad

0.81

icably

0.78

exist

0.77

 bothered

0.76

Activations Density 0.337%

negative sentiments or contrasts within a given context

negations or expressions of absence

No Comments

No Known Activations