INDEX

Explanations

conjunctions

oai_token-act-pair · gpt-3.5-turbo

conjunctions and prepositions indicating relationships or connections between items

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

grave

-0.70

anchester

-0.66

 cardinal

-0.63

 Reilly

-0.63

aganda

-0.59

Jur

-0.58

 Mayer

-0.57

 Tune

-0.57

uberty

-0.55

 dice

-0.54

POSITIVE LOGITS

Accessory

0.82

å§«

0.69

iasm

0.69

çĶ°

0.68

lished

0.65

Cover

0.64

[|

0.63

 intra

0.62

osterone

0.61

ription

0.60

Activations Density 0.034%

conjunctions

conjunctions and prepositions indicating relationships or connections between items

No Comments

No Known Activations