INDEX

Explanations

expressions indicating the desire to investigate or find out more about something

oai_token-act-pair · gpt-3.5-turbo

elements of storytelling and narrative structure

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

icipated

-0.57

erenn

-0.55

 genuinely

-0.54

 culmination

-0.53

 entirety

-0.51

 overwhelmingly

-0.49

ospels

-0.48

 BahÃ¡

-0.48

 mainline

-0.48

chwitz

-0.47

POSITIVE LOGITS

''.

0.70

.?

0.69

.ï¿½

0.68

.''

0.67

`.

0.67

$.

0.66

.(

0.64

_.

0.63

pic

0.63

.#

0.62

Activations Density 1.112%

expressions indicating the desire to investigate or find out more about something

elements of storytelling and narrative structure

No Comments

No Known Activations