INDEX

Explanations

variations of the word "jump" and related actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

 wand

-0.16

cks

-0.15

 Belle

-0.15

allis

-0.15

sein

-0.15

obot

-0.14

pik

-0.14

 Heads

-0.14

 Bronze

-0.14

paged

-0.14

POSITIVE LOGITS

ÏĦÎ®

0.19

frog

0.19

dash

0.17

aroo

0.16

stick

0.16

atron

0.15

ativa

0.15

Jump

0.15

 Jump

0.15

rosse

0.14

Activations Density 0.020%

variations of the word "jump" and related actions

No Comments

No Known Activations