INDEX

Explanations

structured document elements and formatting tags

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.15.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

 Klin

-0.88

albert

-0.76

 Angeles

-0.74

-0.71

Artem

-0.68

al

-0.67

 baker

-0.66

 Richter

-0.65

 albert

-0.65

 Kinney

-0.64

POSITIVE LOGITS

kwargs

1.11

]**

1.10

.**

1.07

(**

0.96

 Monfieur

0.93

'**

0.92

)**

0.91

sizeCache

0.90

:**

0.89

,**

0.87

Activations Density 0.596%

structured document elements and formatting tags

No Comments

No Known Activations