INDEX

Explanations

phrases indicating a range or approximation

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate a range of conditions or variations

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

 corridors

-0.56

Monitor

-0.54

CHR

-0.53

 challeng

-0.53

 Puzz

-0.52

afety

-0.52

 horizont

-0.51

 Vector

-0.51

verages

-0.51

 Traps

-0.50

POSITIVE LOGITS

nery

1.10

 less

0.94

leans

0.93

nam

0.86

gin

0.86

phans

0.84

acular

0.84

chid

0.82

acle

0.80

 fewer

0.77

Activations Density 0.018%

phrases indicating a range or approximation

phrases that indicate a range of conditions or variations

No Comments

No Known Activations