INDEX

Explanations

words related to specific locations or institutions

oai_token-act-pair · gpt-3.5-turbo

proper nouns or names, particularly of people and places

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

arding

-0.79

orate

-0.76

arded

-0.74

iard

-0.71

jamin

-0.70

naissance

-0.70

oard

-0.68

gettable

-0.67

urate

-0.66

umen

-0.66

POSITIVE LOGITS

plings

0.87

ustain

0.85

atchewan

0.84

rano

0.77

pling

0.75

udo

0.75

eways

0.73

Account

0.72

arin

0.70

atoon

0.70

Activations Density 0.144%

words related to specific locations or institutions

proper nouns or names, particularly of people and places

No Comments

No Known Activations