INDEX

Explanations

references to universities and their associated details

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

ened

-0.19

lessly

-0.18

ness

-0.18

 nghiá»ĩp

-0.18

out

-0.16

-0.15

older

-0.15

riday

-0.15

pour

-0.15

nett

-0.15

POSITIVE LOGITS

-wide

0.23

å®Ļ

0.18

wide

0.18

olatile

0.15

ois

0.15

oft

0.15

ATIONAL

0.14

tes

0.14

arians

0.14

(Un

0.14

Activations Density 0.050%

references to universities and their associated details

No Comments

No Known Activations