INDEX

Explanations

specific names and identifiers associated with events, locations, and categorization

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

Viited

-0.79

expandindo

-0.79

 mxArray

-0.78

 nakalista

-0.76

]]

-0.73

Rüyada

-0.68

:])

-0.68

})`

-0.68

 ...]

-0.67

_]

-0.66

POSITIVE LOGITS

 păr

0.60

 specchio

0.57

 penghar

0.55

 giornal

0.53

 NDEBUG

0.52

󠁢

0.52

 kří

0.52

ándome

0.50

Dist

0.50

 spørs

0.48

Activations Density 1.476%

specific names and identifiers associated with events, locations, and categorization

No Comments

No Known Activations