INDEX

Explanations

names and terms related to environmental and community contexts

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

 Reverse

-0.15

disposing

-0.15

umd

-0.14

ERNEL

-0.14

velt

-0.14

akin

-0.13

ochen

-0.13

ilip

-0.13

azen

-0.13

POSITIVE LOGITS

 crisis

0.16

 meiden

0.15

cri

0.14

msp

0.14

mmc

0.14

Ã¼ne

0.14

à¸ŀà¸¢

0.14

atro

0.13

acha

0.13

 recovery

0.13

Activations Density 0.323%

names and terms related to environmental and community contexts

No Comments

No Known Activations