INDEX

Explanations

references to occurrences of the word "first."

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

arkan

-0.18

leaning

-0.14

cul

-0.14

SEL

-0.13

itos

-0.13

avery

-0.13

mekte

-0.13

kol

-0.13

APPED

-0.13

ackson

-0.13

POSITIVE LOGITS

 time

0.66

time

0.42

tim

0.41

.time

0.40

 times

0.39

ti

0.38

	time

0.38

ime

0.38

æĹ¶éĹ´

0.36

-time

0.36

Activations Density 0.029%

references to occurrences of the word "first."

No Known Activations