INDEX

Explanations

names or terms related to persons or places

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

New Auto-Interp

Configuration

jbloom/Gemma-2b-IT-Residual-Stream-SAEs/gemma_2b_it_blocks.12.hook_resid_post_16384

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

HuggingFaceFW/fineweb

Features

16,384

Data Type

float32

Hook Name

blocks.12.hook_resid_post

Hook Layer

Architecture

standard

Context Size

1,024

Dataset

Skylion007/openwebtext

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 NDEBUG

-0.53

Namara

-0.52

 kapag

-0.50

QUIRY

-0.50

werth

-0.49

 Budaya

-0.48

 ajuns

-0.46

 iyong

-0.45

 virtù

-0.45

 Département

-0.44

POSITIVE LOGITS

Rin

0.78

Vin

0.78

Hin

0.76

RIN

0.75

Kin

0.72

Rin

0.72

 Echin

0.72

rin

0.71

Gin

0.71

Cin

0.71

Activations Density 0.307%

names or terms related to persons or places

No Comments

No Known Activations