INDEX

Explanations

occurrences of the word "thread" 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 @ 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

StoryboardSegue

-0.51

templateUrl

-0.44

imy

-0.39

 known

-0.39

omore

-0.38

latimes

-0.37

 borgo

-0.37

⏜

-0.36

fm

-0.36

 القد

-0.36

POSITIVE LOGITS

KURZBESCHREIBUNG

0.71

 Roskov

0.67

 AssemblyProduct

0.66

AddTagHelper

0.64

VersionUID

0.64

>//

0.63

NameInMap

0.63

endpush

0.63

weil

0.63

]")]

0.63

Activations Density 0.018%

occurrences of the word "thread" in various contexts

No Comments

No Known Activations