INDEX

Explanations

phrases related to investigating or looking into various matters or circumstances

oai_token-act-pair · gpt-3.5-turbo

terms related to investigations and inquiries

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

 boasting

-0.67

ffe

-0.64

 dominates

-0.62

 clut

-0.61

 favored

-0.61

 caric

-0.61

throp

-0.60

 boasted

-0.60

 dominated

-0.58

pop

-0.58

POSITIVE LOGITS

 feasibility

0.87

 urgently

0.86

 disciplinary

0.78

 ASAP

0.78

loo

0.78

istrate

0.77

 reconsider

0.77

 damages

0.76

 chall

0.75

 accordingly

0.75

Activations Density 0.650%

phrases related to investigating or looking into various matters or circumstances

terms related to investigations and inquiries

No Comments

No Known Activations