INDEX

Explanations

programming concepts related to logging and error handling

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

static final Logger

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

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_32/width_131k/average_l0_88

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.32.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

出版年

-0.53

ModelAdmin

-0.49

adaptiveStyles

-0.45

 EClass

-0.42

devamını

-0.42

OGND

-0.40

 noDo

-0.39

fjspx

-0.38

routeProvider

-0.38

 ModelExpression

-0.38

POSITIVE LOGITS

log

1.77

 logger

1.72

logger

1.66

Log

1.56

 logging

1.55

 Logger

1.55

 logs

1.55

log

1.55

LOG

1.47

 Logging

1.42

Activations Density 0.045%

programming concepts related to logging and error handling

static final Logger

No Comments

No Known Activations