INDEX

Explanations

references to foot-related activities or objects

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_fs6144-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_6144

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

6,144

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

PsyNetMessage

-0.84

ULE

-0.73

ALD

-0.69

 cannabin

-0.67

effective

-0.66

mble

-0.65

actionDate

-0.64

MRI

-0.64

infeld

-0.64

 Nare

-0.62

POSITIVE LOGITS

prints

1.08

notes

1.08

bridge

1.04

wear

0.95

print

0.94

minster

0.92

steps

0.91

falls

0.90

 stomp

0.90

step

0.90

Activations Density 0.273%

references to foot-related activities or objects

No Comments

No Known Activations