INDEX

Explanations

words related to incentives or disincentives

oai_token-act-pair · gpt-3.5-turbo

words and concepts related to incentives or motivations

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

 incap

-0.67

 answ

-0.66

ŃĶ

-0.65

fac

-0.64

¾

-0.64

 Citiz

-0.64

Fac

-0.60

Ī

-0.59

ipeg

-0.59

Scient

-0.57

POSITIVE LOGITS

hips

0.98

oft

0.96

heet

0.94

hare

0.89

ilver

0.89

cript

0.89

liga

0.85

paces

0.84

terday

0.84

ync

0.82

Activations Density 0.141%

words related to incentives or disincentives

words and concepts related to incentives or motivations

No Comments

No Known Activations