INDEX

Explanations

punctuation marks that indicate excitement or emphasis

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

.ibm

-0.16

 Ð¿Ð¾ÑįÑĤÐ¾Ð¼Ñĥ

-0.15

 Tonight

-0.15

ç»ĻæĪĳ

-0.14

Xxx

-0.14

 duly

-0.14

 consequently

-0.14

GINE

-0.14

ENCE

-0.14

_vlog

-0.14

POSITIVE LOGITS

 another

0.29

 Another

0.25

another

0.24

 Speaking

0.24

 speaking

0.24

Another

0.24

 sometimes

0.23

 although

0.22

 nobody

0.21

 maybe

0.20

Activations Density 0.045%

punctuation marks that indicate excitement or emphasis

No Comments

No Known Activations