INDEX

Explanations

mentions of stacks or arrangements of objects

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>

-1.75

 Francis

-0.55

Ethan

-0.49

private

-0.49

 conduct

-0.48

 James

-0.48

Шо

-0.48

ξ

-0.48

Francis

-0.48

 such

-0.48

POSITIVE LOGITS

 stack

1.34

 stacks

1.30

 Pile

1.30

 Stack

1.28

 pile

1.27

 Stacks

1.23

stack

1.22

 STACK

1.18

 stacking

1.18

Stack

1.17

Activations Density 0.308%

mentions of stacks or arrangements of objects

No Comments

No Known Activations