INDEX

Explanations

mentions of the location "Prospect."

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

endon

-0.17

à¸¡

-0.16

ãĤ¤ãĤº

-0.15

Ã¦k

-0.15

xin

-0.14

esa

-0.14

ÏĦÎ·ÏĥÎ·

-0.14

nd

-0.13

Ves

-0.13

POSITIVE LOGITS

ively

0.31

us

0.22

ors

0.21

pective

0.19

ivism

0.17

ive

0.16

uhan

0.16

uses

0.16

ailles

0.15

itution

0.15

Activations Density 0.012%

mentions of the location "Prospect."

No Comments

No Known Activations