INDEX

Explanations

phrases related to the quality and efficacy of expressions or arguments

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

aleur

-0.15

Ø³ÙħØ©

-0.14

marshall

-0.13

bish

-0.13

lef

-0.13

cts

-0.13

uhl

-0.13

NETWORK

-0.13

ichert

-0.13

ittle

-0.12

POSITIVE LOGITS

 chez

0.14

 ì§Ħíĸī

0.13

emand

0.13

 Goose

0.13

iat

0.13

 Viá»ĩc

0.13

oud

0.13

awi

0.13

resden

0.13

Activations Density 0.038%

phrases related to the quality and efficacy of expressions or arguments

No Comments

No Known Activations