INDEX

Explanations

phrases related to being accustomed to or adapting to new situations

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

<bos>

-2.56

pub

-0.59

 provide

-0.58

 invest

-0.57

 intervene

-0.56

/***

-0.56

 mobilize

-0.56

 engage

-0.55

-0.54

 continue

-0.54

POSITIVE LOGITS

 bandung

1.24

 jacques

1.19

 stockholm

1.17

 ftre

1.17

ftu

1.16

 blackpink

1.15

Juf

1.14

 budapest

1.13

 leonardo

1.13

 bordeaux

1.12

Activations Density 0.496%

phrases related to being accustomed to or adapting to new situations

No Comments

No Known Activations