INDEX

Explanations

occurrences of the word "updated" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

çĶ²

-0.16

rown

-0.15

ä¼

-0.15

-0.14

erez

-0.14

adera

-0.14

pler

-0.14

æĲŃ

-0.13

jo

-0.13

 Genre

-0.13

POSITIVE LOGITS

ysl

0.17

raquo

0.15

-fw

0.15

izzo

0.15

ãģ°ãģĭãĤĬ

0.15

 addCriterion

0.14

Concurrency

0.14

gua

0.14

lsa

0.14

/new

0.14

Activations Density 0.009%

occurrences of the word "updated" in various contexts

No Comments

No Known Activations