INDEX

Explanations

terms related to academic institutions and professional titles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

maal

-0.18

efa

-0.15

aket

-0.15

orts

-0.15

zp

-0.15

 Blink

-0.14

hog

-0.13

meal

-0.13

antly

-0.13

est

-0.13

POSITIVE LOGITS

bourg

0.15

QE

0.15

ourcem

0.15

inati

0.14

emento

0.14

adaÅŁ

0.14

ifter

0.14

 Äĳáº£o

0.14

obil

0.13

Ã¼Ã§

0.13

Activations Density 0.068%

terms related to academic institutions and professional titles

No Comments

No Known Activations