INDEX

Explanations

words related to actions of arriving or entering a location

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "arrived" and related actions

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

bits

-0.73

olin

-0.71

ormal

-0.69

 tongues

-0.66

ãĥª

-0.66

haps

-0.62

 reim

-0.61

 derivatives

-0.60

 implants

-0.59

 bets

-0.58

POSITIVE LOGITS

 peacefully

0.76

 aboard

0.74

 downstairs

0.73

 seiz

0.72

 ashore

0.70

 mosqu

0.70

resp

0.70

 boarded

0.68

 disembark

0.68

 safely

0.66

Activations Density 0.250%

words related to actions of arriving or entering a location

occurrences of the word "arrived" and related actions

No Comments

No Known Activations