INDEX

Explanations

texts related to encouragement and promotion

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>

-3.08

ⓧ

-0.98

-0.95

<?

-0.91

/***

-0.84

/**

-0.80

/*

-0.75

glMatrixMode

-0.74

<?

-0.69

Vegeu

-0.69

POSITIVE LOGITS

 stockholm

1.58

 maroc

1.56

 bandung

1.51

hcm

1.51

 lele

1.48

 wien

1.47

aen

1.46

 meis

1.44

 Minang

1.43

mef

1.42

Activations Density 0.126%

texts related to encouragement and promotion

No Comments

No Known Activations