INDEX

Explanations

concepts related to methods, procedures, and systematic approaches 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

 jScrollPane

-0.48



-0.44

onPressed

-0.42

 lèvres

-0.42

 BoxFit

-0.41

ρων

-0.41

verständlich

-0.41

Aqu

-0.40

maxLines

-0.40

 lagi

-0.39

POSITIVE LOGITS

 described

2.10

 shown

1.97

 mentioned

1.86

 outlined

1.85

 discussed

1.80

 listed

1.80

 depicted

1.74

described

1.74

shown

1.62

 indicated

1.58

Activations Density 2.129%

concepts related to methods, procedures, and systematic approaches in various contexts

No Comments

No Known Activations