INDEX

Explanations

mathematical symbols and equations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

ceae

-0.18

åĿĬ

-0.16

 sims

-0.15

amarin

-0.15

neys

-0.15

 ÙģØ±ÙĪ

-0.15

/GPL

-0.14

áº¬

-0.13

\xaa

-0.13

.Generation

-0.13

POSITIVE LOGITS

 reverse

0.20

 Reverse

0.18

zdy

0.16

Reverse

0.16

reverse

0.16

.tc

0.16

 opposite

0.15

_reverse

0.15

eco

0.15

jeme

0.15

Activations Density 0.053%

mathematical symbols and equations

No Comments

No Known Activations