INDEX

Explanations

curly braces indicating blocks of code or structures

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

oras

-0.16

nim

-0.14

ether

-0.14

aqu

-0.14

izzy

-0.13

éĢ

-0.13

anes

-0.13

 Kahn

-0.13

 Primer

-0.13

_INFORMATION

-0.13

POSITIVE LOGITS

tering

0.16

ghi

0.15

aliz

0.14

anches

0.14

ensa

0.14

UNS

0.14

Ĺı

0.14

 brute

0.14

OwnProperty

0.13

 tatto

0.13

Activations Density 0.004%

curly braces indicating blocks of code or structures

No Comments

No Known Activations