INDEX

Explanations

mentions of specific individuals or 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

Ply

-0.75

 WiFi

-0.70

 stre

-0.69

TERN

-0.65

rities

-0.64

LIN

-0.63

VL

-0.63

LAR

-0.63

 livest

-0.62

 Stellar

-0.61

POSITIVE LOGITS

aq

1.14

af

1.10

ac

1.08

aus

1.05

aed

1.02

ab

1.02

abis

1.01

ax

1.01

amate

1.00

AC

0.99

Activations Density 0.311%

mentions of specific individuals or locations

No Comments

No Known Activations