INDEX

Explanations

phrases that indicate success or notable events

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 11-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_11/width_16k/average_l0_80

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.11.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Monfieur

-0.84

 Theſe

-0.84

 myſelf

-0.74

GrantedAuthority

-0.73

 Paglinawan

-0.72

 Jefus

-0.72

клопе

-0.72

 Shakspeare

-0.71

 Diſ

-0.70

 iſt

-0.69

POSITIVE LOGITS

now

0.61

now

0.53

AndEndTag

0.46

Now

0.44

0.43

<strong>

0.43

Now

0.42

up

0.42

0.41

Activations Density 0.579%

phrases that indicate success or notable events

No Comments

No Known Activations