INDEX

Explanations

references to cycles or repetitive sequences

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

asca

-0.17

igr

-0.17

-0.16

 indefinite

-0.16

lad

-0.15

isto

-0.15

hani

-0.15

Ø¬ÙĪ

-0.15

ame

-0.14

 Morales

-0.14

POSITIVE LOGITS

scoped

0.16

heim

0.16

boats

0.16

kla

0.16

ugins

0.15

istrovstvÃŃ

0.15

bilt

0.15

andin

0.15

cycles

0.15

 gÃ¬

0.15

Activations Density 0.015%

references to cycles or repetitive sequences

No Comments

No Known Activations