INDEX

Explanations

phrases or words related to evaluations or assessments

oai_token-act-pair · gpt-3.5-turbo

references to general assessments or evaluations

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

Caf

-0.82

 Debor

-0.74

rette

-0.71

Bry

-0.67

 Ships

-0.66

 Speedway

-0.66

 Minute

-0.64

 Frie

-0.64

 Ferry

-0.64

 Witnesses

-0.63

POSITIVE LOGITS

 attractiveness

0.81

 stability

0.80

gebra

0.79

heastern

0.79

 averages

0.79

 usefulness

0.76

 overview

0.76

 standings

0.76

 consensus

0.75

 impression

0.75

Activations Density 0.016%

phrases or words related to evaluations or assessments

references to general assessments or evaluations

No Comments

No Known Activations