INDEX

Explanations

names and identifiers related to individuals or entities in conversational contexts

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

 CreateTagHelper

-0.92

 المعيارى

-0.88

 initComponents

-0.80

ostante

-0.77

хьтан

-0.75

Carriera

-0.72

 wireType

-0.71

 Мексичка

-0.69

 décid

-0.69

WriteTagHelper

-0.68

POSITIVE LOGITS

________________

0.47

 Daniel

0.46

^^^^^^^^

0.45

Daniel

0.45

0.43

esi

0.43

tec

0.42

0.41

chers

0.41

0.40

Activations Density 0.124%

names and identifiers related to individuals or entities in conversational contexts

No Comments

No Known Activations