INDEX

Explanations

references to garages

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 1-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_1/width_16k/average_l0_102

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.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

 itſelf

-0.79

 Shakspeare

-0.72

addContainerGap

-0.72

 Majefty

-0.71

 Reſ

-0.71

 esternos

-0.69

 Diſ

-0.68

 themſelves

-0.68

 Chriftian

-0.67

 juſt

-0.67

POSITIVE LOGITS

ParallelGroup

0.91

tex

0.89

TEX

0.84

gg

0.82

Tex

0.74

tex

0.72

FT

0.72

Tex

0.72

TEX

0.69

)++;

0.68

Activations Density 0.057%

references to garages

No Comments

No Known Activations