INDEX

Explanations

references to limits, restrictions, or numerical thresholds, often related to policies or regulations

oai_token-act-pair · gpt-3.5-turbo

references to limits or thresholds, particularly in a financial or regulatory context

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.5.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

ĪĴ

-0.87

hower

-0.74

selves

-0.73

cause

-0.73

cipl

-0.73

isance

-0.68

 Bread

-0.66

VD

-0.66

perse

-0.65

 Roses

-0.63

POSITIVE LOGITS

itol

1.21

aic

0.93

itals

0.87

illary

0.86

rison

0.82

stan

0.78

itated

0.78

aign

0.76

uchin

0.75

acious

0.74

Activations Density 0.012%

references to limits, restrictions, or numerical thresholds, often related to policies or regulations

references to limits or thresholds, particularly in a financial or regulatory context

No Comments

No Known Activations