INDEX

Explanations

names and mentions related to a specific person or character

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_fs3072-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

3,072

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

ãĥ¼ãĥĨãĤ£

-0.82

 sterling

-0.71

olicy

-0.69

 Tycoon

-0.69

 Chaser

-0.65

å£

-0.65

ãģį

-0.64

NT

-0.63

ãģĭ

-0.62

 Solitaire

-0.61

POSITIVE LOGITS

manac

1.33

umni

1.18

gebra

1.16

gorithm

1.13

ibaba

1.11

ameda

1.11

 Jazeera

1.09

oha

1.08

addin

1.07

gorith

1.04

Activations Density 0.329%

names and mentions related to a specific person or character

No Comments

No Known Activations