INDEX

Explanations

statements related to internal processes or operations

oai_token-act-pair · gpt-3.5-turbo

terms related to internal and external processes or actions

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Parenthood

-0.85

 Gamble

-0.79

 Mehran

-0.77

Magikarp

-0.73

EVA

-0.72

 Roses

-0.72

Reply

-0.70

ador

-0.68

Guy

-0.67

eer

-0.65

POSITIVE LOGITS

 displaced

1.02

 internally

1.00

 combustion

0.84

 identifiable

0.81

 speaking

0.80

 exting

0.80

 housed

0.76

vernment

0.75

 dispersed

0.74

 combust

0.74

Activations Density 0.006%

statements related to internal processes or operations

terms related to internal and external processes or actions

No Comments

No Known Activations