INDEX

Explanations

patterns related to shapes and locations

oai_token-act-pair · gpt-3.5-turbo

nouns and terms related to physical elements or constructs 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 @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

 Pigs

-0.71

odox

-0.66

ificantly

-0.65

erous

-0.63

ãĥĥ

-0.63

 guiActiveUn

-0.62

erent

-0.62

amous

-0.59

ighter

-0.58

volent

-0.56

POSITIVE LOGITS

(>

0.81

 surrounded

0.80

 indeed

0.68

(<

0.66

izabeth

0.65

smanship

0.61

 surrounds

0.61

DragonMagazine

0.61

 exists

0.60

cellence

0.60

Activations Density 0.607%

patterns related to shapes and locations

nouns and terms related to physical elements or constructs in various contexts

No Comments

No Known Activations