INDEX

Explanations

words related to the place or event "back"

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

entric

-0.72

viz

-0.66

inational

-0.64

enaries

-0.61

ities

-0.59

Osc

-0.59

ision

-0.58

uary

-0.57

izo

-0.56

ciation

-0.56

POSITIVE LOGITS

packs

1.25

stab

1.17

dated

1.17

tracking

1.14

GROUND

1.13

packing

1.10

lash

1.07

gam

1.04

fires

1.04

door

1.01

Activations Density 0.784%

words related to the place or event "back"

No Comments

No Known Activations