INDEX

Explanations

actions or verbs related to undoing or removing 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.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>

-2.80

/***

-0.79

<?

-0.78

ⓧ

-0.74

<?

-0.69

/**

-0.68

//*/

-0.63

//---

-0.63

 incarcer

-0.59

POSITIVE LOGITS

 lele

1.34

 bandung

1.23

 jawa

1.22

 Minang

1.19

 jati

1.16

 magis

1.10

 jaya

1.04

 saar

1.00

 riva

1.00

 kaos

0.99

Activations Density 0.832%

actions or verbs related to undoing or removing something

No Comments

No Known Activations