INDEX

Explanations

code elements or identifiers related to a specific programming framework or library

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

user prompts

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.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

AndEndTag

-0.88

httphttps

-0.82

 nakalista

-0.74

AddTagHelper

-0.64

ImageContext

-0.59

 Wikimedijinoj

-0.58

الدراسه

-0.57

󠁢

-0.56

ftagPool

-0.56

fromnode

-0.54

POSITIVE LOGITS

tryk

0.33

VisualStyle

0.33

RawData

0.33

ornis

0.33

empuan

0.32

 schéma

0.32

 localization

0.32

 temas

0.32

 Tema

0.31

 Gouvernement

0.31

Activations Density 0.004%

code elements or identifiers related to a specific programming framework or library

user prompts

No Comments

No Known Activations