INDEX

Explanations

proper nouns, like names of people and places

oai_token-act-pair · gpt-3.5-turbo

proper nouns, specifically names of people or characters

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.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

LEASE

-0.70

rium

-0.67

 Corpus

-0.65

berra

-0.65

 Pixie

-0.65

Ñı

-0.65

Lex

-0.64

ktop

-0.61

ÑĮ

-0.60

.ãĢį

-0.59

POSITIVE LOGITS

enegger

1.10

 testified

0.97

 himself

0.90

 wrote

0.84

 told

0.81

 admitted

0.80

 herself

0.80

 remembers

0.80

 admits

0.79

 penned

0.79

Activations Density 0.136%

proper nouns, like names of people and places

proper nouns, specifically names of people or characters

No Comments

No Known Activations