INDEX

Explanations

terms related to parameters and their specifications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

ivery

-0.16

eye

-0.15

OPS

-0.15

ying

-0.15

rops

-0.14

istrat

-0.14

rips

-0.14

eam

-0.14

resses

-0.14

ebb

-0.14

POSITIVE LOGITS

etric

0.43

edics

0.39

etr

0.39

edic

0.37

ilitary

0.36

etrize

0.35

aters

0.33

ters

0.31

etri

0.31

agnetic

0.31

Activations Density 0.017%

terms related to parameters and their specifications

No Comments

No Known Activations