INDEX

Explanations

technical terms and well-structured information

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

New Auto-Interp

Configuration

jbloom/Gemma-2b-Residual-Stream-SAEs/gemma_2b_blocks.6.hook_resid_post_16384_anthropic_fast_lr

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

HuggingFaceFW/fineweb

Features

16,384

Data Type

torch.float32

Hook Point

blocks.6.hook_resid_post

Architecture

standard

Context Size

1,024

Dataset

HuggingFaceFW/fineweb

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

<bos>

-3.21

-0.91

/***

-0.83

ⓧ

-0.81

<?

-0.67

ScopeManager

-0.65

<?

-0.63

principalTable

-0.62

addCriterion

-0.62

/**

-0.62

POSITIVE LOGITS

">...

1.04

 véhic

1.04

 soulign

1.03

lijah

0.94

dison

0.88

 roul

0.88

âteau

0.87

 étr

0.86

 catég

0.86

 prétend

0.86

Activations Density 0.049%

technical terms and well-structured information

No Comments

No Known Activations