INDEX

Explanations

words related to gadgets and technology products

oai_token-act-pair · gpt-3.5-turbo

mentions of specific individuals or entities, particularly names associated with events or contexts

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

âķĲ

-0.82

Ļ½

-0.82

anwhile

-0.75

cean

-0.74

terday

-0.73

ylum

-0.70

lander

-0.70

Ëľ

-0.70

 Larson

-0.69

ï¸

-0.67

POSITIVE LOGITS

gets

0.96

ijah

0.89

entary

0.85

arov

0.82

estones

0.81

ahon

0.79

afi

0.78

rop

0.78

obook

0.77

onga

0.77

Activations Density 0.033%

words related to gadgets and technology products

mentions of specific individuals or entities, particularly names associated with events or contexts

No Comments

No Known Activations