INDEX

Explanations

phrases related to negative issues and challenges in various contexts

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

<bos>

-2.13

/**

-1.04

ⓧ

-0.93

<?

-0.91

<?

-0.87

/*

-0.84

public

-0.80

#![

-0.79

/***

-0.78

POSITIVE LOGITS

 jaya

1.73

aen

1.67

 stockholm

1.67

 maroc

1.65

 bandung

1.65

 maneu

1.65

hcm

1.65

 dises

1.64

 affor

1.64

 wien

1.63

Activations Density 0.164%

phrases related to negative issues and challenges in various contexts

No Known Activations