INDEX

Explanations

mentions of specific and personal names

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

ATION

-0.76

enhagen

-0.73

âĸ¬âĸ¬

-0.66

ATIONS

-0.66

ITED

-0.65

ations

-0.63

 clearance

-0.63

atories

-0.62

 Clover

-0.61

sto

-0.61

POSITIVE LOGITS

ttp

1.29

urst

1.16

orns

1.13

ulhu

1.10

irteen

1.10

ouse

1.08

reshold

1.07

orne

1.06

ought

1.05

awk

1.05

Activations Density 2.233%

mentions of specific and personal names

No Comments

No Known Activations