INDEX

Explanations

names of individuals associated with influential contributions, events, or contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

werfen

-0.45

malıdır

-0.42

str

-0.42

ตก

-0.41

griffen

-0.41

em

-0.41

หมด

-0.40

tritte

-0.40

 setbacks

-0.40

fälle

-0.40

POSITIVE LOGITS

AddTagHelper

0.60



0.58

GOTREF

0.55

complexContent

0.54

},

0.54

WriteBarrier

0.51

ագրություններ

0.51

 Мексичка

0.50

[],

0.50

Referanser

0.50

Activations Density 0.052%

names of individuals associated with influential contributions, events, or contexts

No Comments

No Known Activations