INDEX

Explanations

the word "no" with a high activation value

oai_token-act-pair · gpt-3.5-turbo

phrases expressing the concept of "none" or "no."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

RAFT

-0.72

endar

-0.72

 towed

-0.70

adobe

-0.66

 bombed

-0.65

 reborn

-0.64

 tossed

-0.62

 Lear

-0.60

 Cutter

-0.60

 hone

-0.58

POSITIVE LOGITS

xious

1.14

oses

1.00

except

0.99

 longer

0.95

 doubt

0.90

 avail

0.87

oooooooo

0.86

 whatsoever

0.83

obs

0.81

oooooooooooooooo

0.81

Activations Density 0.087%

the word "no" with a high activation value

phrases expressing the concept of "none" or "no."

No Comments

No Known Activations