INDEX

Explanations

phrases related to taking action or making progress

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

-Ta

-0.16

indsight

-0.15

ongan

-0.15

zon

-0.15

erdale

-0.14

road

-0.14

lems

-0.14

osc

-0.14

erk

-0.14

 smoke

-0.14

POSITIVE LOGITS

ilar

0.18

0.16

Klo

0.16

 taken

0.15

wizard

0.15

atur

0.15

 ÐŁÐ»Ð¾

0.15

pery

0.14

 Taken

0.14

éª

0.14

Activations Density 0.015%

phrases related to taking action or making progress

No Comments

No Known Activations