INDEX

Explanations

references to relevance and pertinent information 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 @ 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

LayoutPanel

-0.80

extAlignment

-0.75

ers

-0.73

AnchorStyles

-0.64

BeginContext

-0.63

Deposits

-0.62

-0.60

Fns

-0.60

bå

-0.59

TypedDataSet

-0.59

POSITIVE LOGITS

LEVANT

1.25

levance

1.09

 Relevance

1.00

Relevant

1.00

Relevance

1.00

 Relevant

0.90

relevant

0.88

 relevant

0.87

 Irrelevant

0.86

 pertin

0.86

Activations Density 0.065%

references to relevance and pertinent information in various contexts

No Comments

No Known Activations