INDEX

Explanations

references to individual names and entities within 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 @ 10-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_10/width_16k/average_l0_77

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.10.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

 olímp

-0.53

ủa

-0.52

hæng

-0.51

GEBURTS

-0.50

何况

-0.50

enance

-0.50

".

-0.49

 traverse

-0.49

untura

-0.48

POSITIVE LOGITS

:+:

0.93

 propOrder

0.71

 GenerationType

0.68

expandindo

0.62

 CreateTagHelper

0.61

abestanden

0.60

PhysRev

0.57

resave

0.56

Rhestr

0.56

 isEmpty

0.54

Activations Density 0.366%

references to individual names and entities within a context

No Comments

No Known Activations