INDEX

Explanations

connective terms that suggest enhancement or improvement 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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

MigrationBuilder

-0.70

AutoScaleMode

-0.61

 estekak

-0.60

 ſche

-0.60

 listo

-0.58

 صوتيه

-0.56

 lode

-0.56

SharedCtor

-0.55

 indemnity

-0.55

 Freeway

-0.55

POSITIVE LOGITS

 also

0.64

enderror

0.59

she

0.58

reportWebVitals

0.57

 còn

0.55

She

0.54

0.53

 connexes

0.53

 Chwiliwch

0.53

RUnlock

0.51

Activations Density 0.318%

connective terms that suggest enhancement or improvement in various contexts

No Comments

No Known Activations