INDEX

Explanations

references to support networks and relationships in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 оригіналу

-0.92

quelize

-0.77

PerformLayout

-0.73

 CreateTagHelper

-0.68

rawDesc

-0.66

AndEndTag

-0.66

ViewImports

-0.65

itself

-0.65

 iconTwitter

-0.64

Manbalar

-0.63

POSITIVE LOGITS

who

1.12

 whom

1.00

 capable

0.96

 capables

0.93

who

0.81

whom

0.77

capable

0.73

 willing

0.72

 capaces

0.72

 able

0.71

Activations Density 0.352%

references to support networks and relationships in various contexts

No Comments

No Known Activations