INDEX

Explanations

words and phrases associated with goal-oriented actions or outcomes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 38-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_38/width_16k/average_l0_128

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.38.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 ujednoznacz

-0.45

 ולה

-0.43

OGND

-0.43

NewLabel

-0.42

 OFDb

-0.41

 Chriftian

-0.40

ſp

-0.40

 fubject

-0.40

 fevere

-0.39

 perfons

-0.39

POSITIVE LOGITS

 itself

0.87

tifies

0.76

itself

0.75

us

0.71

ulates

0.69

izes

0.68

 nothing

0.67

cribes

0.66

pires

0.65

 Itself

0.62

Activations Density 0.917%

words and phrases associated with goal-oriented actions or outcomes

No Comments

No Known Activations