INDEX

Explanations

structured data elements and their attributes within a coding context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

ighet

-0.16

oji

-0.15

 Ð¶Ðµ

-0.15

çĶ²

-0.15

oa

-0.15

igh

-0.15

suming

-0.14

issing

-0.14

Ã£

-0.13

Ø±Ø§ÙĤ

-0.13

POSITIVE LOGITS

sw

0.15

èĬĻ

0.15

nete

0.15

eshire

0.15

.Fat

0.15

ailles

0.14

unsch

0.14

lags

0.14

istration

0.14

ibraries

0.13

Activations Density 0.019%

structured data elements and their attributes within a coding context

No Comments

No Known Activations