INDEX

Explanations

terms related to enhancement or improvement

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 10-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_10/width_16k/average_l0_77

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.10.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

 absolut

-0.58

grpc

-0.58

 nothing

-0.57

rag

-0.57

 Tsche

-0.56

inaw

-0.56

sy

-0.56

ILayout

-0.55

plotly

-0.55

 wide

-0.55

POSITIVE LOGITS

 deepened

1.05

 deepen

1.04

 deepening

1.04

ENED

0.99

 Strengthen

0.96

strengthen

0.94

Strengthen

0.93

ValueStyle

0.93

 sharpen

0.92

 strengthen

0.90

Activations Density 0.352%

terms related to enhancement or improvement

No Comments

No Known Activations