INDEX

Explanations

information related to exceeding a certain amount or threshold

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

IFrame

Link

Negative Logits

Vegeu

-0.73

<bos>

-0.65

RouterModule

-0.63

FileInputStream

-0.63

Referències

-0.62

///**

-0.61

 continue

-0.59

Requisitos

-0.59

ляю

-0.59

Література

-0.59

POSITIVE LOGITS

 Excess

2.88

 excess

2.85

excess

2.71

Excess

2.46

 excesses

2.02

 exceso

1.69

 surplus

1.68

 excès

1.59

 Surplus

1.56

 Excessive

1.52

Activations Density 0.125%

information related to exceeding a certain amount or threshold

No Known Activations