INDEX

Explanations

dates and time-related expressions

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

áº¯m

-0.16

 Trou

-0.15

 kabil

-0.15

ÏĥÎ±

-0.15

osas

-0.14

lech

-0.14

brtc

-0.14

POSITIVE LOGITS

0.79

Û±Û¹Û´

0.51

0.47

0.35

Û±Û¹Ûµ

0.31

 WWII

0.24

0.22

 Stalin

0.19

 wartime

0.18

Activations Density 0.055%

dates and time-related expressions

No Comments

No Known Activations