INDEX

Explanations

significant numeric or alphanumeric identifiers and terms related to measurement or specification

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

ÄĽr

-0.18

ustos

-0.15

ekte

-0.15

keh

-0.15

íĭĢ

-0.14

OLER

-0.14

 genie

-0.14

templ

-0.14

ceph

-0.14

ëŁ

-0.14

POSITIVE LOGITS

ive

0.17

0.16

uu

0.15

ival

0.15

èį

0.15

cu

0.15

0.14

INET

0.14

ip

0.14

ation

0.14

Activations Density 0.033%

significant numeric or alphanumeric identifiers and terms related to measurement or specification

No Comments

No Known Activations