INDEX

Explanations

first-person references and expressions of personal experience or inquiry

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

edd

-0.15

Ð¦Ðµ

-0.15

unami

-0.14

Hop

-0.14

 Gott

-0.14

okit

-0.14

esco

-0.14

wright

-0.14

 Hopkins

-0.14

ague

-0.14

POSITIVE LOGITS

ow

0.16

dash

0.15

dub

0.15

 nghá»ĭ

0.15

herits

0.14

Jub

0.14

0.13

ç³»

0.13

Activations Density 0.024%

first-person references and expressions of personal experience or inquiry

No Comments

No Known Activations