INDEX

Explanations

phrases related to affirmations

oai_token-act-pair · gpt-3.5-turbo

affirmations or confirmations within the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

abal

-0.77

èĢħ

-0.76

arted

-0.73

rance

-0.70

ridor

-0.69

ounded

-0.66

vati

-0.65

liner

-0.64

Discussion

-0.64

gall

-0.64

POSITIVE LOGITS

sir

0.96

 technically

0.79

 THERE

0.77

yes

0.74

 please

0.69

 anecd

0.67

 yeah

0.66

 sexism

0.65

 there

0.64

 insofar

0.63

Activations Density 0.050%

phrases related to affirmations

affirmations or confirmations within the text

No Comments

No Known Activations