INDEX

Explanations

instances of the word "match" or variations thereof

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 4-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_4/width_16k/average_l0_124

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.4.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

quirrel

-0.81

zzleHttp

-0.76

бенок

-0.74

thâu

-0.72

skyl

-0.70

 kasarigan

-0.70

::-

-0.70

hehe

-0.69

Vader

-0.69



-0.69

POSITIVE LOGITS

 match

2.63

 Match

2.58

 MATCH

2.58

Match

2.50

match

2.50

 matches

2.44

MATCH

2.32

 Matches

2.29

matches

2.06

Matches

2.02

Activations Density 0.050%

instances of the word "match" or variations thereof

No Comments

No Known Activations