INDEX

Explanations

phrases related to events, updates, and notifications

oai_token-act-pair · gpt-3.5-turbo

references to outcomes and responses related to investigations or inquiries

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 staple

-0.68

 paradise

-0.67

 ageing

-0.67

 nursery

-0.62

 invincible

-0.62

 planner

-0.61

 mimic

-0.60

 envy

-0.59

 forgetting

-0.59

 beck

-0.59

POSITIVE LOGITS

displayText

0.90

 tonight

0.89

 regarding

0.84

Thursday

0.81

.#

0.80

 here

0.79

Tuesday

0.76

Wednesday

0.75

 herein

0.75

 JUSTICE

0.75

Activations Density 0.310%

phrases related to events, updates, and notifications

references to outcomes and responses related to investigations or inquiries

No Comments

No Known Activations