INDEX

Explanations

terms related to evaluating risks and making decisions based on beliefs or convictions

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Configuration

jbloom/Gemma-2b-IT-Residual-Stream-SAEs/gemma_2b_it_blocks.12.hook_resid_post_16384

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

HuggingFaceFW/fineweb

Features

16,384

Data Type

float32

Hook Name

blocks.12.hook_resid_post

Hook Layer

Architecture

standard

Context Size

1,024

Dataset

Skylion007/openwebtext

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

coeffs

-0.65

 Gies

-0.60

Be

-0.58

 Hermans

-0.57

 Kunt

-0.57

freqs

-0.56

 Backman

-0.55

 Knud

-0.55

She

-0.55

idxs

-0.55

POSITIVE LOGITS

 affez

1.53

 sappi

1.39

 soggior

1.36

 ristor

1.32

<bos>

1.32

 vogli

1.24

 abbra

1.23

 Ottobre

1.19

 apparti

1.18

 tranquillo

1.18

Activations Density 0.734%

terms related to evaluating risks and making decisions based on beliefs or convictions

No Comments

No Known Activations