INDEX

Explanations

instances of nested structures or blocks in programming syntax

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 27-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L27R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.27.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

EIF

-0.15

ãĥĪãĥ«

-0.14

ØªØ¯

-0.14

Ø³Ø§Ùħ

-0.14

_partial

-0.14

å¯

-0.13

ÐºÐ°ÑĢ

-0.13

ogany

-0.13

ichni

-0.13

ynchron

-0.13

POSITIVE LOGITS

0.22

ajo

0.16

rx

0.16

aggi

0.15

enu

0.15

ENU

0.15

 å°

0.15

ÂłÂłÂłÂłÂłÂłÂł

0.15

ox

0.15

iano

0.14

Activations Density 0.089%

instances of nested structures or blocks in programming syntax

No Comments

No Known Activations