INDEX

Explanations

expressions related to the understanding or fairness of a situation

oai_token-act-pair · gpt-3.5-turbo

expressions of justification or reasonableness

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

 bowling

-0.77

 downed

-0.67

worm

-0.65

craft

-0.61

 virginity

-0.61

ngth

-0.60

bows

-0.59

infect

-0.59

Dur

-0.59

stars

-0.58

POSITIVE LOGITS

 assume

0.72

 inference

0.71

ensibly

0.69

 assumption

0.68

ATOR

0.68

 conclude

0.67

allery

0.67

 infer

0.66

 Luxem

0.66

 guiActive

0.66

Activations Density 0.139%

expressions related to the understanding or fairness of a situation

expressions of justification or reasonableness

No Comments

No Known Activations