INDEX

Explanations

names and titles of individuals as well as entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

ihar

-0.16

.operations

-0.15

esis

-0.15

angler

-0.15

rir

-0.15

TEGER

-0.15

probante

-0.14

PackageManager

-0.14

POSITIVE LOGITS

 extinct

0.15

 Rossi

0.14

eren

0.14

 Pare

0.14

nop

0.14

 Nash

0.14

azzi

0.13

onec

0.13

UZ

0.13

¥

0.13

Activations Density 0.051%

names and titles of individuals as well as entities

No Comments

No Known Activations