INDEX

Explanations

words related to errors and feedback in a coding context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.15.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

 ModelExpression

-0.85

setVerticalGroup

-0.81

 nakalista

-0.69

StoreMessageInfo

-0.68

UnsafeEnabled

-0.66

 Audiodateien

-0.62

 &___

-0.62

addCriterion

-0.61

ArrowToggle

-0.61

 مشين

-0.60

POSITIVE LOGITS

 preſent

0.68

 ſche

0.66

 purpoſe

0.64

 ſtate

0.64

 pleaſure

0.64

 houſe

0.63

 ſmall

0.61

 perſon

0.59

 myſelf

0.57

 Conſ

0.57

Activations Density 0.083%

words related to errors and feedback in a coding context

No Comments

No Known Activations