INDEX

Explanations

identifiers and class names in programming code

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

<start_of_turn>

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

SequentialGroup

-0.66

 nahilalakip

-0.65

 <>",

-0.65

writeFieldEnd

-0.62

 protoimpl

-0.61

 nakalista

-0.59

principalColumn

-0.59

errHandler

-0.57

ondissement

-0.57

 ſever

-0.55

POSITIVE LOGITS

XmlRootElement

0.41

AutoScaleMode

0.40

class

0.34

 Miche

0.34

spo

0.33

Tcp

0.33

 Taylor

0.32

 siff

0.32

Dr

0.31

btnNew

0.31

Activations Density 0.007%

identifiers and class names in programming code

<start_of_turn>

No Comments

No Known Activations