INDEX

Explanations

expressions of interest or concern

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 0-gemmascope-mlp-65k

Configuration

google/gemma-scope-2b-pt-mlp/layer_0/width_65k/average_l0_72

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.0.hook_mlp_out

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

 متعلقه

-0.90

RenderAtEndOf

-0.88

хьтан

-0.88

__*/

-0.86

uxxxx

-0.79

AccessorTable

-0.78

ViewFeatures

-0.75

 समीक्षक

-0.74

 AssemblyProduct

-0.73

IsContent

-0.73

POSITIVE LOGITS

 talked

0.65

 discussed

0.57

 fought

0.56

combination

0.56

0.55

 advocated

0.52

 debated

0.52

NetworkInfo

0.52

optarg

0.52

球

0.51

Activations Density 0.037%

expressions of interest or concern

No Comments

No Known Activations