INDEX

Explanations

phrases that express a sense of belonging or connection

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L1R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.1.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

mund

-0.16

.SIG

-0.15

Ð¾ÑĢÐ°

-0.15

CKET

-0.15

iry

-0.14

ANTE

-0.14

stile

-0.14

 Affairs

-0.14

icz

-0.14

Å¯l

-0.14

POSITIVE LOGITS

0.16

anuts

0.15

xec

0.14

Toolkit

0.14

cro

0.14

ense

0.14

enos

0.13

 ensuing

0.13

itler

0.13

 resident

0.13

Activations Density 0.032%

phrases that express a sense of belonging or connection

No Comments

No Known Activations