INDEX

Explanations

terms related to sleep and rest

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

ãĥ£

-0.71

 Robertson

-0.69

McN

-0.66

cture

-0.64

govtrack

-0.64

izons

-0.63

 Manson

-0.63

æł

-0.62

tein

-0.61

 radial

-0.61

POSITIVE LOGITS

walking

1.04

wake

1.00

leep

0.96

walker

0.94

Sleep

0.87

 sleeper

0.85

walk

0.85

mares

0.84

mare

0.84

aby

0.82

Activations Density 0.956%

terms related to sleep and rest

No Comments

No Known Activations