INDEX

Explanations

programming constructs and method definitions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_21/width_16k/average_l0_70

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

aarrggbb

-0.77

if

-0.51

 litté

-0.44

db

-0.42

 attore

-0.42

assert

-0.41

global

-0.39

 mandatario

-0.39

 waivers

-0.39

 Reminis

-0.39

POSITIVE LOGITS

 public

0.98

public

0.91

protected

0.82

 onOptions

0.81

AccessorTable

0.79

DockStyle

0.79

AnchorStyles

0.78

parsedMessage

0.76

AndEndTag

0.75

complexContent

0.74

Activations Density 0.055%

programming constructs and method definitions

No Comments

No Known Activations