INDEX

Explanations

terms related to insensitivity or disregard

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

New Auto-Interp

Configuration

fnlp/Llama3_1-8B-Base-LXM-8x/Llama3_1-8B-Base-L22M-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.22.hook_mlp_out

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

othy

-0.17

imuth

-0.16

ivery

-0.15

icina

-0.15

pton

-0.15

.DropTable

-0.14

otionEvent

-0.14

edBy

-0.14

nal

-0.14

kdir

-0.14

POSITIVE LOGITS

ins

0.20

urn

0.18

peg

0.17

ulet

0.17

ulin

0.16

 shapes

0.15

ins

0.15

ens

0.15

upa

0.15

Ø¨Ø§ØŃ

0.15

Activations Density 0.013%

terms related to insensitivity or disregard

No Comments

No Known Activations