INDEX

Explanations

words related to playful or mysterious hints and clues

oai_token-act-pair · gpt-3.5-turbo

instances of the word "tease" and its variations

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

gur

-0.75

sen

-0.74

rage

-0.73

ternity

-0.72

nea

-0.71

imm

-0.70

ski

-0.69

dan

-0.68

itars

-0.67

pert

-0.66

POSITIVE LOGITS

 tease

1.01

 hints

1.00

 tantal

0.94

 clues

0.88

 hint

0.86

 teasing

0.78

 glimps

0.76

 teaser

0.73

 appet

0.70

 teased

0.69

Activations Density 0.030%

words related to playful or mysterious hints and clues

instances of the word "tease" and its variations

No Comments

No Known Activations