INDEX

Explanations

words related to customer feedback about websites or products

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

<bos>

-3.70

 Distrikt

-0.96

-0.94

RuleContext

-0.93

<th>

-0.92

util

-0.91

/**

-0.90

—

-0.89

/**

-0.89

//

-0.89

POSITIVE LOGITS

 reluct

3.61

 affor

3.49

 increa

3.48

 shenan

3.48

 maneu

3.41

 disagre

3.41

 impra

3.33

 depic

3.29

 unlaw

3.24

 unwarran

3.18

Activations Density 6.964%

words related to customer feedback about websites or products

No Comments

No Known Activations