INDEX

Explanations

the word "outlook"

oai_token-act-pair · gpt-4-turbo

mentions or discussions about future potential or expectations

oai_token-act-pair · gpt-3.5-turbo

references to future predictions or assessments

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

cing

-0.80

Interstitial

-0.77

pmwiki

-0.75

ced

-0.74

adding

-0.69

avis

-0.69

tein

-0.68

 Explos

-0.67

Anth

-0.67

Lago

-0.64

POSITIVE LOGITS

 outlook

1.42

 horizon

0.87

eus

0.85

aution

0.75

 forecasts

0.73

 Outlook

0.73

 downgrade

0.72

eering

0.72

 trajectory

0.72

igible

0.72

Activations Density 0.008%

the word "outlook"

mentions or discussions about future potential or expectations

references to future predictions or assessments

No Comments

No Known Activations