INDEX

Explanations

phrases indicating the presence and availability of features or components in products

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.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

uch

-0.15

pta

-0.15

çĶ±

-0.15

 match

-0.15

wen

-0.14

son

-0.14

jc

-0.14

GIN

-0.14

aml

-0.14

POSITIVE LOGITS

olini

0.15

Lon

0.15

itore

0.15

âĸį

0.14

 Sala

0.14

lÃ¡n

0.14

Laf

0.14

LOPT

0.14

anic

0.14

andi

0.14

Activations Density 0.072%

phrases indicating the presence and availability of features or components in products

No Comments

No Known Activations