INDEX

Explanations

abstract concepts or goals

oai_token-act-pair · gpt-3.5-turbo

concepts and terms related to objectives or claims

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

ctors

-0.67

umbn

-0.67

idth

-0.63

utters

-0.61

idates

-0.61

headers

-0.60

ãĥ³ãĤ¸

-0.60

 condem

-0.59

srf

-0.59

usha

-0.57

POSITIVE LOGITS

 ourselves

0.95

 myself

0.90

 firsthand

0.82

.<

0.76

 yourself

0.75

 unconsciously

0.73

 vividly

0.73

 empir

0.72

ality

0.71

 manually

0.71

Activations Density 0.240%

abstract concepts or goals

concepts and terms related to objectives or claims

No Comments

No Known Activations