INDEX

Explanations

mentions of locations or establishments, mostly related to restaurants or entertainment venues

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

Not in Any Lists

No Comments

Negative Logits

<bos>

-1.37

ുറ

-0.83

ⓧ

-0.78

public

-0.77

enumerate

-0.77

///**

-0.75

-0.74

IEnumerator

-0.73

void

-0.73

/***

-0.73

POSITIVE LOGITS

 maneu

2.08

 stockholm

2.02

 affor

1.99

 increa

1.91

 wien

1.84

 lidl

1.84

 accla

1.81

 inev

1.81

 impra

1.79

 eiffel

1.79

Activations Density 0.101%

mentions of locations or establishments, mostly related to restaurants or entertainment venues

No Comments

No Known Activations