INDEX

Explanations

references to structured or systematic components in a context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

 "..\..\..\

-0.69

UnusedPrivate

-0.69

 disambiguazione

-0.69

AddTagHelper

-0.62

 odkazy

-0.62

PreferredItem

-0.61

<bos>

-0.60

脚注の使い方

-0.57

'{@

-0.52

tvguidetime

-0.52

POSITIVE LOGITS

▼

0.70

anair

0.69

 Mitch

0.64

Marne

0.61

❋

0.61

Allora

0.61

 Haben

0.58

pytest

0.57

 typelib

0.56

--}}

0.55

Activations Density 0.057%

references to structured or systematic components in a context

No Comments

No Known Activations