INDEX

Explanations

names of people or characters with a possessive form

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>

-3.83

/**

-0.99

/*

-0.93

<?

-0.92

/***

-0.89

ⓧ

-0.87

-0.85

Vegeu

-0.79

/*++

-0.78

 AssemblyCompany

-0.76

POSITIVE LOGITS

 affor

1.93

 increa

1.92

 unlaw

1.90

Juf

1.85

 reluct

1.84

 impra

1.82

 wherea

1.79

 volunte

1.78

 disagre

1.75

 maneu

1.75

Activations Density 0.424%

names of people or characters with a possessive form

No Comments

No Known Activations