INDEX

Explanations

the word "inbox" with high activation values, indicating a focus on email-related content and communication

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>

-1.49

public

-0.67

glBind

-0.66

protected

-0.66

 realize

-0.66

fileprivate

-0.61

/**

-0.61

glLoadIdentity

-0.61

glBegin

-0.60

POSITIVE LOGITS

aen

1.53

 erec

1.50

 accla

1.50

 noel

1.48

 casio

1.45

 maroc

1.45

 impra

1.44

 effe

1.43

 wien

1.43

 dises

1.41

Activations Density 0.105%

the word "inbox" with high activation values, indicating a focus on email-related content and communication

No Comments

No Known Activations

the word "inbox" with high activation values, indicating a focus on email-related content and communication

No Comments

No Known Activations