INDEX

Explanations

text related to reports, statements, and communication between individuals or entities

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs24576-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 doom

-0.82

 charisma

-0.76

ãĤ¼ãĤ¦ãĤ¹

-0.76

 mound

-0.73

 sidx

-0.72

 landslide

-0.72

llular

-0.69

 ãĤµãĥ¼ãĥĨãĤ£ãĥ¯ãĥ³

-0.67

erella

-0.65

 fert

-0.64

POSITIVE LOGITS

Statement

0.84

 apologised

0.82

"#

0.81

 refunds

0.80

 spokeswoman

0.80

 clarification

0.79

 discontinued

0.78

hement

0.78

 complied

0.76

 declined

0.76

Activations Density 0.466%

text related to reports, statements, and communication between individuals or entities

No Comments

No Known Activations