INDEX

Explanations

words related to brains

oai_token-act-pair · gpt-3.5-turbo

the concept of "gains" or positive outcomes in various contexts

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

 Kinnikuman

-0.77

enhagen

-0.72

Ens

-0.70

mercial

-0.69

ophon

-0.69

govtrack

-0.69

olog

-0.68

 skelet

-0.67

album

-0.67

nec

-0.67

POSITIVE LOGITS

hene

0.76

ains

0.73

IFT

0.71

pring

0.70

cott

0.68

fo

0.68

wic

0.68

ees

0.67

igue

0.66

ugal

0.65

Activations Density 0.018%

words related to brains

the concept of "gains" or positive outcomes in various contexts

No Comments

No Known Activations