INDEX

Explanations

programming-related tokens, particularly those associated with function definitions and runtime operations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

Ð²Ð¾Ð»

-0.15

 Mediterranean

-0.15

IRD

-0.14

andi

-0.14

ramid

-0.14

.hasMore

-0.14

ä»¿

-0.14

.Xml

-0.13

SP

-0.13

 Chicago

-0.13

POSITIVE LOGITS

Nan

0.26

<v

0.23

NAN

0.23

(isolate

0.21

JS

0.20

 isolate

0.20

js

0.19

.cljs

0.18

 JavaScript

0.18

tpl

0.17

Activations Density 0.001%

programming-related tokens, particularly those associated with function definitions and runtime operations

No Comments

No Known Activations