INDEX

Explanations

instances of structured data and object-related terminology

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 23-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_23/width_16k/average_l0_75

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.23.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

</i>

-1.45

</code>

-0.97

</h6>

-0.91

</b>

-0.87

</td>

-0.86

</th>

-0.86

&#

-0.79

-0.77

<b>

-0.71

-0.69

POSITIVE LOGITS

adaptiveStyles

1.35

SourceChecksum

1.21

AddTagHelper

1.11

tagHelperRunner

1.10

ſelves

1.08

 vooz

1.04

يكب

1.04

𑄮

1.04

ſelf

1.02

الحياه

1.01

Activations Density 0.018%

instances of structured data and object-related terminology

No Comments

No Known Activations