INDEX

Explanations

specific names of films or characters

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

edin

-0.17

pler

-0.15

ikh

-0.15

Ã¡y

-0.15

yla

-0.15

olid

-0.14

ellar

-0.14

iente

-0.14

chez

-0.14

DetailsService

-0.14

POSITIVE LOGITS

Î»ÎµÎ¹

0.16

áº¯c

0.14

æĪ²

0.14

rott

0.14

 catalogs

0.14

æľºåħ³

0.13

 Sommer

0.13

ROWS

0.13

tti

0.13

Ð¾Ð±Ð¸

0.13

Activations Density 0.054%

specific names of films or characters

No Comments

No Known Activations