INDEX

Explanations

instances of the word "to" and other related expressions indicating intention or purpose

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

stery

-0.14

keterangan

-0.14

ned

-0.14

nip

-0.13

munition

-0.13

 batching

-0.13

Ã±as

-0.13

Detach

-0.12

.Ui

-0.12

uish

-0.12

POSITIVE LOGITS

arrow

0.15

$MESS

0.13

.asp

0.13

oad

0.13

_UNSIGNED

0.13

ä»²

0.13

.PARAM

0.13

Å¯l

0.12

hem

0.12

_make

0.12

Activations Density 0.726%

instances of the word "to" and other related expressions indicating intention or purpose

No Comments

No Known Activations