INDEX

Explanations

references to the name "Aaron."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

-0.17

tain

-0.16

gar

-0.15

operative

-0.14

opping

-0.14

dal

-0.14

ied

-0.14

reh

-0.14

POSITIVE LOGITS

son

0.21

 Rodgers

0.21

 Neville

0.17

 Burr

0.16

tru

0.16

 Judge

0.16

sons

0.15

swith

0.15

ewan

0.15

alysis

0.15

Activations Density 0.004%

references to the name "Aaron."

No Comments

No Known Activations