INDEX

Explanations

references to specific characters and their traits in text

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

ego

-0.17

etary

-0.17

egin

-0.17

.sharedInstance

-0.17

erton

-0.17

ary

-0.16

enberg

-0.16

enas

-0.15

air

-0.15

ery

-0.15

POSITIVE LOGITS

istically

0.25

isation

0.20

ised

0.19

izations

0.18

pent

0.18

med

0.17

untime

0.17

oque

0.16

nels

0.16

ually

0.16

Activations Density 0.049%

references to specific characters and their traits in text

No Comments

No Known Activations