INDEX

Explanations

symbols and formatting elements typically found in digital or HTML content

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

oble

-0.15

Åº

-0.15

IZER

-0.15

outu

-0.14

-CS

-0.14

ichel

-0.14

esson

-0.14

otel

-0.13

AGER

-0.13

oller

-0.13

POSITIVE LOGITS

ãĥŃãĥ³

0.15

iken

0.14

unc

0.14

wis

0.14

Kit

0.14

_exempt

0.14

plode

0.14

wish

0.14

.sex

0.14

intColor

0.14

Activations Density 0.002%

symbols and formatting elements typically found in digital or HTML content

No Comments

No Known Activations