INDEX

Explanations

occurrences of the word "Filed" often indicating categorization or archival of content

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

swire

-0.15

patial

-0.15

orz

-0.15

 Sabb

-0.14

orsk

-0.14

avern

-0.14

oru

-0.13

CASCADE

-0.13

erable

-0.13

abi

-0.13

POSITIVE LOGITS

llen

0.15

obot

0.15

esh

0.15

afort

0.15

lington

0.14

.want

0.14

isay

0.14

 isNew

0.14

wand

0.14

endir

0.14

Activations Density 0.002%

occurrences of the word "Filed" often indicating categorization or archival of content

No Comments

No Known Activations