INDEX

Explanations

fragments of code or programming elements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 5-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L5R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.5.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

emb

-0.16

angs

-0.15

_PRIORITY

-0.14

 hÃ¡t

-0.14

izo

-0.14

ang

-0.13

 normalized

-0.13

ch

-0.13

POSITIVE LOGITS

ODEV

0.16

_INLINE

0.15

nist

0.15

odon

0.15

 Kostenlose

0.15

ndef

0.14

uctor

0.14

_allocator

0.14

reib

0.13

sumer

0.13

Activations Density 0.058%

fragments of code or programming elements

No Comments

No Known Activations