INDEX

Explanations

words related to enhancements or improvements

oai_token-act-pair · gpt-3.5-turbo

terms related to enhancements and reductions in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

Ring

-0.84

Eth

-0.83

abet

-0.82

rics

-0.77

Asia

-0.76

çī

-0.75

NOT

-0.75

GP

-0.74

goal

-0.73

Shop

-0.70

POSITIVE LOGITS

 satell

1.04

 carbohyd

1.00

DeliveryDate

0.92

undai

0.92

ĸļå£«

0.92

 subur

0.92

 millenn

0.92

nesday

0.90

 confir

0.88

 versions

0.87

Activations Density 0.051%

words related to enhancements or improvements

terms related to enhancements and reductions in various contexts

No Comments

No Known Activations