INDEX

Explanations

references to application programming concepts and framework elements in code snippets

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

‘

-0.56

-0.53

 pieds

-0.46

--)

-0.46

“

-0.46

mag

-0.45

ra

-0.44

 True

-0.44

-0.43

encodeWith

-0.43

POSITIVE LOGITS

 pleaſure

0.81

AndEndTag

0.76

 reaſon

0.71

ScopeManager

0.70

 Houſe

0.70

 houſe

0.69

Util

0.69

Utils

0.69

TagHelper

0.68

 ſaid

0.68

Activations Density 0.145%

references to application programming concepts and framework elements in code snippets

No Comments

No Known Activations