INDEX

Explanations

references to specific people or entities, particularly related to sports and culture

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 33-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_33/width_16k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.33.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

AddTagHelper

-0.54

errHandler

-0.53

SequentialGroup

-0.52

 MessageBoxIcon

-0.51

aarrggbb

-0.49

 CURIAM

-0.49

PYX

-0.48

HasAnnotation

-0.47

awtextra

-0.47

 flèche

-0.47

POSITIVE LOGITS

Pil

0.79

Pel

0.72

pil

0.71

Pel

0.70

Pil

0.70

PEL

0.68

pel

0.66

PIL

0.65

pil

0.64

 pilo

0.60

Activations Density 0.173%

references to specific people or entities, particularly related to sports and culture

No Comments

No Known Activations