INDEX

Explanations

specific information that is valuable or helpful

oai_token-act-pair · gpt-3.5-turbo

references to "tips" in various 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

Zed

-0.69

CN

-0.68

 NAME

-0.67

ruciating

-0.67

 Asylum

-0.66

NCT

-0.66

effic

-0.66

Ago

-0.66

 Nationwide

-0.65

ITNESS

-0.65

POSITIVE LOGITS

tip

1.25

tip

1.19

 tips

0.95

jar

0.95

heet

0.87

 toes

0.87

ster

0.86

tips

0.85

Tip

0.85

Tip

0.84

Activations Density 0.022%

specific information that is valuable or helpful

references to "tips" in various contexts

No Comments

No Known Activations