INDEX

Explanations

information about incidents or events happening at specific locations

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

nir

-0.84

ledged

-0.80

©¶æ¥µ

-0.73

ItemTracker

-0.68

Ĥİ

-0.67

²¾

-0.65

intendent

-0.63

afort

-0.63

isma

-0.63

Ħ¢

-0.62

POSITIVE LOGITS

 unfold

1.44

 videos

1.09

 replay

1.08

 VIDEOS

1.02

 tapes

1.01

 video

0.99

 closely

0.94

 movies

0.94

 movie

0.93

 clip

0.93

Activations Density 0.143%

information about incidents or events happening at specific locations

No Comments

No Known Activations