INDEX

Explanations

questions related to reasons or explanations

oai_token-act-pair · gpt-3.5-turbo

interrogative phrases that seek explanations or clarifications

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

boa

-0.70

 slightest

-0.66

gi

-0.64

Interstitial

-0.64

fm

-0.62

ONSORED

-0.62

ĸļ

-0.62

aciously

-0.62

Ire

-0.61

andro

-0.61

POSITIVE LOGITS

 Makes

0.93

 Difference

0.92

 Does

0.92

 Documents

0.88

 Benefits

0.86

 Facts

0.86

 Factors

0.85

 Definitions

0.85

 Uses

0.84

Own

0.83

Activations Density 0.138%

questions related to reasons or explanations

interrogative phrases that seek explanations or clarifications

No Comments

No Known Activations