INDEX

Explanations

references to AMD module definitions and related 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 @ 6-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L6R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.6.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

inia

-0.16

fixtures

-0.15

ãģ«ãģĭ

-0.15

ssi

-0.15

enburg

-0.14

perse

-0.14

_wp

-0.14

getic

-0.14

.MixedReality

-0.14

Ð¾Ð·Ð²

-0.14

POSITIVE LOGITS

ãĥ¼ãĤ¯

0.16

oplay

0.15

Ear

0.15

ZONE

0.15

IOD

0.15

ÙĬØ§Øª

0.14

.lt

0.14

asses

0.14

 Blank

0.14

Activations Density 0.370%

references to AMD module definitions and related programming concepts

No Comments

No Known Activations