INDEX

Explanations

references to specific data types or programming structures

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

Writes

-0.14

 Ã¢

-0.14

ÑĢÐ°Ð

-0.14

Äįet

-0.13

linkplain

-0.13

laps

-0.13

 processData

-0.12

à¤Ĺà¤²

-0.12

.Footer

-0.12

 kÃ¢

-0.12

POSITIVE LOGITS

 AssemblyCopyright

0.17

 âĢª

0.16

&↵

0.14

Ì£

0.14

 :↵↵↵↵

0.14

 ÎĳÏĢ

0.14

 :-↵

0.13

ATAL

0.13

ÂŃing

0.13

Activations Density 1.258%

references to specific data types or programming structures

No Comments

No Known Activations