INDEX

Explanations

phrases that describe situations or conditions

oai_token-act-pair · gpt-3.5-turbo

the word "where" in the context of various scenarios or situations

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

ÂłÂłÂłÂłÂłÂłÂłÂł

-0.68

 Doctrine

-0.60

rolet

-0.60

apult

-0.59

 Touch

-0.58

 Beasts

-0.57

ÂłÂłÂł

-0.57

 Thief

-0.57

 Beetle

-0.56

wu

-0.56

POSITIVE LOGITS

upon

1.41

soever

1.25

abouts

1.20

fore

1.07

with

0.77

earch

0.73

ãĥ¯ãĥ³

0.70

rency

0.68

ħ

0.64

 they

0.64

Activations Density 0.061%

phrases that describe situations or conditions

the word "where" in the context of various scenarios or situations

No Comments

No Known Activations