INDEX

Explanations

words related to providing helpful suggestions or feedback

oai_token-act-pair · gpt-3.5-turbo

terms related to constructive actions and feedback, often in contrast to abusive or negative behavior

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

orph

-0.90

atch

-0.80

urg

-0.76

ared

-0.75

aver

-0.72

gars

-0.72

alm

-0.69

olog

-0.69

paralle

-0.69

andra

-0.68

POSITIVE LOGITS

 constructive

1.15

 criticism

0.85

 feedback

0.83

-+-+

0.77

 entreprene

0.76

 redes

0.75

 repr

0.74

 sunlight

0.71

 daylight

0.71

 criticisms

0.70

Activations Density 0.011%

words related to providing helpful suggestions or feedback

terms related to constructive actions and feedback, often in contrast to abusive or negative behavior

No Comments

No Known Activations