INDEX

Explanations

keywords related to reasoning, justification, or argumentation based on certain conditions or principles

oai_token-act-pair · gpt-3.5-turbo

phrases that refer to justification or reasoning for actions or opinions

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

ascus

-0.79

jer

-0.77

ortium

-0.75

apore

-0.73

aspers

-0.71

chuk

-0.70

ravis

-0.69

torn

-0.68

nova

-0.67

uckland

-0.67

POSITIVE LOGITS

 basis

0.89

 assumption

0.88

 grounds

0.82

 occasions

0.77

 premise

0.76

 criteria

0.76

 principles

0.75

 whim

0.73

 behalf

0.71

 clauses

0.71

Activations Density 0.054%

keywords related to reasoning, justification, or argumentation based on certain conditions or principles

phrases that refer to justification or reasoning for actions or opinions

No Comments

No Known Activations