INDEX

Explanations

phrases indicating probability or likelihood

oai_token-act-pair · gpt-3.5-turbo

phrases expressing likelihood or probability

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

ciating

-0.74

arantine

-0.73

perty

-0.72

ilan

-0.70

GBT

-0.70

elight

-0.70

nai

-0.69

Ô

-0.68

rency

-0.68

gall

-0.68

POSITIVE LOGITS

 that

1.20

 there

0.86

 they

0.85

 THAT

0.84

that

0.75

 whoever

0.74

we

0.72

 none

0.70

 many

0.68

she

0.68

Activations Density 0.191%

phrases indicating probability or likelihood

phrases expressing likelihood or probability

No Comments

No Known Activations