INDEX

Explanations

references to political processes and nominations

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

Ð´Ð°Ð²

-0.16

oola

-0.16

acier

-0.14

uyu

-0.14

ole

-0.13

abb

-0.13

 ä¹

-0.13

AtPath

-0.13

hausen

-0.13

olest

-0.13

POSITIVE LOGITS

 floor

0.29

 debate

0.29

 Floor

0.25

-floor

0.24

 clot

0.24

 roll

0.23

 motion

0.23

 Debate

0.23

Floor

0.23

floor

0.23

Activations Density 0.029%

references to political processes and nominations

No Comments

No Known Activations