INDEX

Explanations

proper nouns related to people's names

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "rev" and terms related to altitude or elevation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

--------------------------------------------------------

-0.67

cknowled

-0.65

 Flags

-0.63

ed

-0.62

 Terr

-0.59

ß

-0.58

 Barn

-0.58

Bl

-0.58

 Views

-0.57

 Knot

-0.57

POSITIVE LOGITS

tenance

0.98

amins

0.89

stery

0.89

ardless

0.88

ythm

0.85

ridor

0.85

dden

0.85

ãĤ£

0.84

andise

0.83

artist

0.83

Activations Density 0.042%

proper nouns related to people's names

occurrences of the word "rev" and terms related to altitude or elevation

No Comments

No Known Activations