INDEX

Explanations

names of locations or place names

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 22-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L22R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.22.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

elta

-0.15

ILON

-0.15

illo

-0.15

amel

-0.14

odÃŃ

-0.14

aru

-0.14

rada

-0.14

ãİ

-0.14

cÃŃ

-0.14

CREMENT

-0.14

POSITIVE LOGITS

oslav

0.17

rrha

0.16

halt

0.15

osate

0.14

 partially

0.14

íĥĦ

0.14

ë²Į

0.14

PUS

0.14

opoulos

0.14

 halt

0.13

Activations Density 0.086%

names of locations or place names

No Comments

No Known Activations