INDEX

Explanations

written reviews

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Configuration

jbloom/Gemma-2b-Residual-Stream-SAEs/gemma_2b_blocks.0.hook_resid_post_16384_anthropic

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

HuggingFaceFW/fineweb

Features

16,384

Data Type

torch.float32

Hook Point

blocks.0.hook_resid_post

Architecture

standard

Context Size

1,024

Dataset

HuggingFaceFW/fineweb

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Gorb

-0.83

Juf

-0.76

 Schrö

-0.69

 Rine

-0.68

 mépris

-0.66

 Bartholo

-0.63

 Idem

-0.62

 Ibidem

-0.59

ⓧ

-0.57

 HAROLD

-0.57

POSITIVE LOGITS

Reviewed

3.43

 Reviewed

1.75

reviewed

1.21

Reviewer

0.88

InjectAttribute

0.76

 reviewed

0.74

 ajustable

0.69

 استنادى

0.68

 Reviewer

0.63

Sucesor

0.61

Activations Density 0.171%

written reviews

No Comments

No Known Activations