INDEX

Explanations

phrases related to assigning responsibility or fault to someone or something

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

<bos>

-1.27

public

-0.69

protected

-0.69

Vegeu

-0.68

export

-0.67

sec

-0.65

/**

-0.64

</thead>

-0.63

生

-0.62

/**

-0.62

POSITIVE LOGITS

 blame

2.53

 Blame

2.43

blame

2.18

 blames

2.13

 blaming

1.98

 stockholm

1.95

 blamed

1.95

Blame

1.91

 accla

1.87

 maneu

1.86

Activations Density 0.161%

phrases related to assigning responsibility or fault to someone or something

No Comments

No Known Activations