INDEX

Explanations

comments or statements made by individuals

oai_token-act-pair · gpt-3.5-turbo

instances of the word "comment" or its variations indicating opinions or statements made by individuals

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

nown

-0.93

 Sinai

-0.85

 turf

-0.74

ãĥ¼ãĥ«

-0.70

married

-0.68

lay

-0.68

adena

-0.68

riz

-0.67

 Lansing

-0.67

fold

-0.65

POSITIVE LOGITS

ariat

0.96

ature

0.86

ively

0.84

 favorably

0.82

ault

0.76

ivity

0.76

atures

0.75

atively

0.74

ivating

0.74

ivated

0.73

Activations Density 0.025%

comments or statements made by individuals

instances of the word "comment" or its variations indicating opinions or statements made by individuals

No Comments

No Known Activations