INDEX

Explanations

expressions or phrases indicating significant events or transitions

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

ses

-0.19

iffe

-0.15

usive

-0.14

abeth

-0.13

xbf

-0.13

ceae

-0.13

á»ĥ

-0.13

aments

-0.13

feas

-0.13

oul

-0.13

POSITIVE LOGITS

erif

0.18

er

0.15

entifier

0.15

0.14

UPPORTED

0.14

ohana

0.14

ingham

0.14

entication

0.14

––

0.14

jed

0.14

Activations Density 0.101%

expressions or phrases indicating significant events or transitions

No Comments

No Known Activations