INDEX

Explanations

names or terms related to individuals, potentially with some emphasis or importance

oai_token-act-pair · gpt-3.5-turbo

references to specific people and their relationships or affiliations

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

ãĤ´ãĥ³

-0.74

Ô

-0.72

 envy

-0.68

 responsiveness

-0.67

CLASS

-0.67

ItemThumbnailImage

-0.65

ãĤ¤ãĥĪ

-0.65

wards

-0.64

 heels

-0.64

 srfAttach

-0.62

POSITIVE LOGITS

pora

0.98

liction

0.92

nih

0.87

ritical

0.75

ect

0.71

ritic

0.71

cephal

0.71

ĳå£«

0.70

OUS

0.70

etus

0.69

Activations Density 0.067%

names or terms related to individuals, potentially with some emphasis or importance

references to specific people and their relationships or affiliations

No Comments

No Known Activations