INDEX

Explanations

phrases related to approval, appreciation, and recognition

oai_token-act-pair · gpt-3.5-turbo

references to public recognition and support

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

 Neuroscience

-0.67

 Tycoon

-0.62

umbledore

-0.60

estamp

-0.59

iru

-0.59

 Abstract

-0.59

 tomorrow

-0.58

 Intermediate

-0.57

ogens

-0.55

 Begin

-0.54

POSITIVE LOGITS

 lately

1.35

 since

1.03

 steadily

1.00

 recent

0.90

 sporadic

0.88

 consistently

0.87

 spor

0.86

 numerous

0.83

since

0.81

Since

0.75

Activations Density 1.206%

phrases related to approval, appreciation, and recognition

references to public recognition and support

No Comments

No Known Activations