INDEX

Explanations

words related to remaining or continuing in a particular place or situation

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_fs1536-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

1,536

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

 Cosponsors

-0.93

DragonMagazine

-0.72

sha

-0.72

oros

-0.65

ramid

-0.65

NetMessage

-0.65

onica

-0.62

furt

-0.62

ologies

-0.62

kson

-0.61

POSITIVE LOGITS

cation

1.07

 afloat

1.06

 intact

1.03

 unchanged

0.99

 indefinitely

0.89

 steadfast

0.84

 undefeated

0.84

emate

0.83

 untouched

0.83

 unnoticed

0.82

Activations Density 1.310%

words related to remaining or continuing in a particular place or situation

No Comments

No Known Activations