INDEX

Explanations

comparisons and conditional expressions related to programming concepts

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

illet

-0.16

åŁºåľ°

-0.15

pose

-0.15

ordion

-0.15

ÐµÐ´Ð°Ð³

-0.15

é§

-0.15

iddle

-0.14

trainer

-0.14

puted

-0.14

villa

-0.14

POSITIVE LOGITS

smith

0.17

awn

0.16

mov

0.16

 College

0.15

COL

0.15

 Moody

0.14

Ð°ÑĢÐ¾Ð²

0.14

cons

0.14

 Cons

0.14

 Carb

0.14

Activations Density 0.027%

comparisons and conditional expressions related to programming concepts

No Comments

No Known Activations