INDEX

Explanations

programming terms related to traits and implementations

oai_token-act-pair · gpt-3.5-turbo

mentions of "traits" and their attributes 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 @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

©¶æ

-0.77

vernment

-0.74

 Copenhagen

-0.73

gran

-0.70

 Lank

-0.70

aii

-0.69

cision

-0.65

Ã±

-0.63

DEL

-0.63

 Arkham

-0.63

POSITIVE LOGITS

 traits

1.33

 trait

1.26

 behaviours

0.97

rait

0.96

ively

0.81

ologies

0.80

 behaviors

0.80

 genes

0.79

 predis

0.78

ouri

0.77

Activations Density 0.007%

programming terms related to traits and implementations

mentions of "traits" and their attributes in various contexts

No Comments

No Known Activations