INDEX

Explanations

phrases related to specific actions or steps in instructions

oai_token-act-pair · gpt-3.5-turbo

references to fans and related activities or terminology

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

.","

-0.66

 ..."

-0.63

.</

-0.63

OTA

-0.62

toggle

-0.62

..."

-0.62

Enlarge

-0.61

,...

-0.60

---

-0.59

âĢ

-0.57

POSITIVE LOGITS

theless

0.84

intosh

0.82

 smoker

0.78

miah

0.74

strous

0.72

uterte

0.71

etheless

0.70

Dialogue

0.70

ertodd

0.69

 gore

0.69

Activations Density 0.385%

phrases related to specific actions or steps in instructions

references to fans and related activities or terminology

No Comments

No Known Activations