INDEX

Explanations

words related to warnings and alerts

oai_token-act-pair · gpt-3.5-turbo

words and phrases related to analysis and insights, particularly in a questioning or investigative context

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

ouver

-0.72

aux

-0.67

 docks

-0.65

kin

-0.64

 contiguous

-0.63

 quart

-0.63

 halls

-0.62

 pounds

-0.61

 exhibits

-0.61

 cart

-0.60

POSITIVE LOGITS

ingly

0.81

inducing

0.76

ustration

0.74

oggle

0.73

terday

0.71

hift

0.71

 worrying

0.71

faced

0.69

bh

0.68

ably

0.67

Activations Density 0.252%

words related to warnings and alerts

words and phrases related to analysis and insights, particularly in a questioning or investigative context

No Comments

No Known Activations