INDEX

Explanations

references to specific names, particularly related to people or entities

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

leşti

-0.53

těte

-0.50

 demain

-0.50

 McIntosh

-0.46

 Poitiers

-0.46

Contexts

-0.46

le

-0.46

mailto

-0.45

icoot

-0.45

 ceci

-0.44

POSITIVE LOGITS

 للمعارف

0.88

interopRequire

0.79

发表于

0.74

]")]

0.65

IsPostBack

0.63

 invokingState

0.63

InteropServices

0.63

__).

0.63

<>());

0.63

disposing

0.61

Activations Density 0.087%

references to specific names, particularly related to people or entities

No Comments

No Known Activations