INDEX

Explanations

symbols and punctuation elements used in marking section breaks or figures

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

hea

-0.14

↵

-0.14

 overst

-0.14

 mostly

-0.13

asm

-0.13

ãĥ¼ãĥŃ

-0.13

 tent

-0.13

stra

-0.13

BI

-0.13

POSITIVE LOGITS

istrovstvÃŃ

0.16

ÄįÃŃ

0.16

ADDE

0.15

lius

0.15

erdale

0.14

 modne

0.14

:UITableView

0.14

 addCriterion

0.14

ÙİØ§ÙĨ

0.13

ednou

0.13

Activations Density 0.036%

symbols and punctuation elements used in marking section breaks or figures

No Comments

No Known Activations