INDEX

Explanations

phrases starting with "One" followed by a statement or observation

oai_token-act-pair · gpt-3.5-turbo

the repeated use of the word "One" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

hips

-0.88

osponsors

-0.77

 actionGroup

-0.73

="#

-0.69

ÃįÃį

-0.69

uits

-0.62

eworld

-0.62

ships

-0.62

berus

-0.62

ãĤ¦

-0.61

POSITIVE LOGITS

 hundred

1.13

 thing

1.11

 Hundred

1.01

 drawback

0.98

 caveat

0.96

 Piece

0.95

 reason

0.94

 aspect

0.90

 notable

0.88

 wonders

0.87

Activations Density 0.069%

phrases starting with "One" followed by a statement or observation

the repeated use of the word "One" in various contexts

No Comments

No Known Activations