INDEX

Explanations

phrases where the word "ask" is closely associated with "for"

oai_token-act-pair · gpt-3.5-turbo

phrases that involve requesting or asking for assistance

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

edin

-0.74

port

-0.74

alde

-0.73

uded

-0.66

iltr

-0.64

oller

-0.64

anch

-0.64

alth

-0.62

 Stru

-0.62

edge

-0.61

POSITIVE LOGITS

 forgiveness

1.18

 clarification

0.99

 permission

0.96

 refunds

0.88

 reinforcements

0.86

gotten

0.82

 inclusion

0.82

gery

0.80

 donations

0.79

 ransom

0.79

Activations Density 0.067%

phrases where the word "ask" is closely associated with "for"

phrases that involve requesting or asking for assistance

No Comments

No Known Activations