INDEX

Explanations

instances of the word "to" in varying contexts, suggesting a focus on purpose or intent

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 19-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_19/width_16k/average_l0_73

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.19.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

 TextAppearance

-0.82

onnaissance

-0.75

CodeAttribute

-0.74

 متعلقه

-0.73

EndContext

-0.72

StoryboardSegue

-0.71

ίδι

-0.68

 виправивши

-0.67

yled

-0.66

INCT

-0.65

POSITIVE LOGITS

me

0.89

 having

0.67

Having

0.56

 Having

0.55

 have

0.53

 suddenly

0.52

him

0.50

 faptul

0.49

 rius

0.48

had

0.48

Activations Density 0.172%

instances of the word "to" in varying contexts, suggesting a focus on purpose or intent

No Comments

No Known Activations