INDEX

Explanations

references to specific numbers or codes, possibly related to technical or administrative contexts

oai_token-act-pair · gpt-3.5-turbo

references to specific numerical codes or identifiers

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

liga

-0.65

ardless

-0.62

redit

-0.61

shape

-0.60

anooga

-0.59

 Phill

-0.58

 Clancy

-0.58

velt

-0.57

isol

-0.57

spread

-0.57

POSITIVE LOGITS

0.97

th

0.78

0.75

0.72

 Sins

0.72

0.70

0.68

mph

0.67

0.66

nm

0.65

Activations Density 0.035%

references to specific numbers or codes, possibly related to technical or administrative contexts

references to specific numerical codes or identifiers

No Comments

No Known Activations