INDEX

Explanations

terms related to mechanical components and switches

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

arden

-0.16

tun

-0.14

 Mobility

-0.14

hel

-0.14

 tuning

-0.13

unnel

-0.13

fft

-0.13

 mobility

-0.13

_singleton

-0.13

Jac

-0.13

POSITIVE LOGITS

 bist

0.23

 relay

0.23

 logic

0.22

logic

0.21

Logic

0.21

 Relay

0.20

 trip

0.20

 Logic

0.20

 energ

0.19

 trips

0.19

Activations Density 0.065%

terms related to mechanical components and switches

No Comments

No Known Activations