INDEX

Explanations

HTML tags, particularly those associated with bold formatting

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

Tokens preceding specific HTML tags

np_acts-logits-general · gemini-2.0-flash

app settings and options

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_16k/average_l0_76

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.31.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

CppMethod

-0.65

SequentialGroup

-0.64

::_('

-0.53

addPreferredGap

-0.51

mybatisplus

-0.50

RTHOOK

-0.49

Hochspringen

-0.49

="@+

-0.46

aspectj

-0.46

:✨

-0.45

POSITIVE LOGITS

‘

0.68

0.55

<em>

0.52

 című

0.51

 يتيمه

0.48

“)

0.47

nın

0.47

<b>

0.47

samt

0.47

️

0.46

Activations Density 0.266%

HTML tags, particularly those associated with bold formatting

Tokens preceding specific HTML tags

app settings and options

No Comments

No Known Activations