INDEX

Explanations

references to historical accuracy and authenticity in religious texts

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

 reader

-0.17

Gri

-0.16

reader

-0.16

WithDuration

-0.15

etsk

-0.14

pert

-0.14

.mult

-0.14

 polis

-0.14

oir

-0.13

ew

-0.13

POSITIVE LOGITS

ipop

0.18

ãĥªãĥ³ãĤ°

0.18

å±

0.15

 Maher

0.15

0.14

 tÃ´

0.14

TextWriter

0.14

 Literary

0.14

æŁ»

0.14

 Cycle

0.14

Activations Density 0.107%

references to historical accuracy and authenticity in religious texts

No Comments

No Known Activations