INDEX

Explanations

phrases indicating lack of evidence or support for claims

oai_token-act-pair · gpt-3.5-turbo

phrases indicating the presence or absence of evidence supporting claims

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

hots

-0.75

 wastes

-0.74

iasis

-0.72

got

-0.69

rolled

-0.64

ches

-0.63

 Needs

-0.63

 typo

-0.62

abad

-0.61

 Nurs

-0.61

POSITIVE LOGITS

 justify

1.29

 suggest

1.25

 substant

1.21

 indicate

1.20

 prove

1.12

 refute

1.11

 validate

1.09

 imply

1.06

 warrant

1.05

 contradict

1.03

Activations Density 0.087%

phrases indicating lack of evidence or support for claims

phrases indicating the presence or absence of evidence supporting claims

No Comments

No Known Activations