INDEX

Explanations

terms related to graph theory and structure

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-mlp-65k

Configuration

google/gemma-scope-2b-pt-mlp/layer_18/width_65k/average_l0_88

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.18.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

########.

-0.55

sys

-0.49

 splinter

-0.48

WindowConstants

-0.47

 question

-0.47

 Oxygen

-0.46

 link

-0.44

 oxygen

-0.43

geben

-0.42

 rent

-0.41

POSITIVE LOGITS

aarrggbb

0.74

httphttps

0.73

 resourceCulture

0.72

 Monfieur

0.67

Démographie

0.66

 ModelExpression

0.66

 doubtnut

0.65

SOUNDBITE

0.63

)");

0.61

)"),

0.61

Activations Density 0.344%

terms related to graph theory and structure

No Comments

No Known Activations