INDEX

Explanations

references to programming concepts and technical terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 7-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L7R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.7.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

Sto

-0.16

sto

-0.15

urf

-0.15

ored

-0.15

ents

-0.14

rote

-0.14

 Angus

-0.14

agn

-0.14

ep

-0.14

epam

-0.14

POSITIVE LOGITS

wner

0.16

undry

0.16

locks

0.16

ailles

0.16

ÅĻÃŃd

0.15

æĥł

0.14

plen

0.14

vez

0.14

Activations Density 0.021%

references to programming concepts and technical terms

No Comments

No Known Activations