INDEX

Explanations

sentence boundaries in the text

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

 ―――――

-0.88

 ProtoMessage

-0.84

Efq

-0.82

Tikang

-0.76

 itſelf

-0.74

 gyhoeddwyd

-0.74

 فريبيس

-0.73

 chng

-0.71

 becauſe

-0.71

 myſelf

-0.71

POSITIVE LOGITS

0.54

idosis

0.53

 vulgares

0.52

0.51

 riso

0.51

AreEqual

0.49

 general

0.49

خلص

0.49

 Craig

0.49

rá

0.48

Activations Density 0.009%

sentence boundaries in the text

No Comments

No Known Activations