INDEX

Explanations

names and words related to specific individuals or locations

oai_token-act-pair · gpt-3.5-turbo

the repeated mention of specific names, particularly "Wald," indicating a focus on notable figures or entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

SER

-0.69

 partName

-0.67

EEE

-0.64

INTER

-0.63

EED

-0.61

BILITY

-0.60

 dism

-0.60

ICES

-0.59

rael

-0.58

 Occupations

-0.58

POSITIVE LOGITS

stad

1.04

stadt

1.02

orf

0.97

nesday

0.93

 Wald

0.92

heim

0.91

roth

0.90

enstein

0.88

rop

0.87

©¶æ¥µ

0.87

Activations Density 0.009%

names and words related to specific individuals or locations

the repeated mention of specific names, particularly "Wald," indicating a focus on notable figures or entities

No Comments

No Known Activations