INDEX

Explanations

terms related to championships and competitive events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 5-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L5R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.5.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

íı°

-0.18

iances

-0.15

ying

-0.15

ileÅŁ

-0.14

IES

-0.14

thing

-0.14

habit

-0.14

Fuse

-0.14

HEL

-0.14

POSITIVE LOGITS

rists

0.17

rav

0.16

ry

0.15

rys

0.15

cy

0.15

edl

0.15

okus

0.14

stag

0.14

wy

0.14

iox

0.14

Activations Density 0.010%

terms related to championships and competitive events

No Comments

No Known Activations