INDEX

Explanations

adjectives describing strong beliefs or characteristics

oai_token-act-pair · gpt-3.5-turbo

terms that signal strong beliefs or positions

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

daq

-0.82

REDACTED

-0.77

iership

-0.71

Mind

-0.69

oping

-0.68

nesota

-0.68

iosity

-0.65

xiety

-0.64

ither

-0.61

Report

-0.60

POSITIVE LOGITS

glers

0.85

ãĤ¡

0.83

ï¿½ï¿½ï¿½ï¿½

0.76

char

0.76

aneers

0.75

sonian

0.74

\\\\

0.72

ãĥİ

0.71

essential

0.71

ãĤ©

0.70

Activations Density 0.058%

adjectives describing strong beliefs or characteristics

terms that signal strong beliefs or positions

No Comments

No Known Activations