INDEX

Explanations

phrases with unusual characters, potentially indicating errors in text encoding or formatting

oai_token-act-pair · gpt-3.5-turbo

the repeated occurrences of a specific character or symbol

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

 orgasm

-0.57

 precon

-0.57

 deduction

-0.57

 imagination

-0.57

 misunder

-0.57

 gist

-0.57

 objective

-0.55

 successes

-0.55

 optimizations

-0.55

 pleasures

-0.54

POSITIVE LOGITS

ï¸ı

1.21

conom

0.80

ï¸

0.76

0.75

gypt

0.72

taboola

0.71

âĢł

0.71

oil

0.67

cffffcc

0.66

Â½

0.65

Activations Density 0.615%

phrases with unusual characters, potentially indicating errors in text encoding or formatting

the repeated occurrences of a specific character or symbol

No Comments

No Known Activations