INDEX

Explanations

interrogative words or questions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

irth

-0.18

GenerationStrategy

-0.16

ognito

-0.15

bole

-0.15

rier

-0.15

BuilderFactory

-0.14

Äįen

-0.14

ingo

-0.14

pta

-0.14

ullan

-0.14

POSITIVE LOGITS

ãĥįãĥ«

0.16

wil

0.15

achel

0.14

dee

0.14

åį°

0.14

icer

0.14

wil

0.13

ILLE

0.13

@qq

0.13

Activations Density 0.027%

interrogative words or questions

No Comments

No Known Activations