INDEX

Explanations

mentions of awards and nominations

oai_token-act-pair · gpt-3.5-turbo

references to awards and nominations

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

chy

-0.77

heter

-0.75

gypt

-0.75

IGHTS

-0.72

peror

-0.71

 Seym

-0.70

tics

-0.70

ipt

-0.70

 inval

-0.70

adra

-0.69

POSITIVE LOGITS

seller

1.06

iary

0.95

sell

0.91

 Selling

0.86

 Worst

0.81

iaries

0.80

selling

0.80

 Best

0.79

 Practices

0.78

 BaseType

0.77

Activations Density 0.031%

mentions of awards and nominations

references to awards and nominations

No Comments

No Known Activations