INDEX

Explanations

expressions indicating a certain sentiment or attitude towards a situation or statement

oai_token-act-pair · gpt-3.5-turbo

expressions indicating familiarity or mediocrity

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

oulos

-0.86

perty

-0.72

edia

-0.72

 VIDEOS

-0.69

IBLE

-0.68

oppers

-0.64

KS

-0.62

ilitary

-0.62

å§«

-0.62

çīĪ

-0.61

POSITIVE LOGITS

este

0.74

¹

0.69

cast

0.69

nered

0.68

 grapes

0.68

hearted

0.67

etter

0.65

ling

0.64

lier

0.64

assy

0.64

Activations Density 0.026%

expressions indicating a certain sentiment or attitude towards a situation or statement

expressions indicating familiarity or mediocrity

No Comments

No Known Activations