INDEX

Explanations

information related to events that have occurred or actions taken by individuals

oai_token-act-pair · gpt-3.5-turbo

phrases that introduce or reference an incident or event

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

ulp

-0.66

 VIDEOS

-0.65

Â¢

-0.65

 Solution

-0.62

nor

-0.61

Plus

-0.60

cery

-0.60

TT

-0.60

die

-0.59

prime

-0.59

POSITIVE LOGITS

 lasted

0.98

 comprises

0.93

 consisted

0.93

 consists

0.90

 prompted

0.89

 prompts

0.87

 culmin

0.85

 amounted

0.84

 occurred

0.83

 lasts

0.82

Activations Density 0.116%

information related to events that have occurred or actions taken by individuals

phrases that introduce or reference an incident or event

No Comments

No Known Activations