INDEX

Explanations

words related to heightened states or intensification

oai_token-act-pair · gpt-3.5-turbo

terms related to heightened states of awareness or concern

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

phis

-0.96

ica

-0.83

icate

-0.77

phia

-0.76

ocide

-0.74

strate

-0.73

usive

-0.72

icates

-0.69

gered

-0.68

ļéĨĴ

-0.68

POSITIVE LOGITS

renheit

0.80

voy

0.78

rons

0.77

mare

0.74

chio

0.73

RIS

0.68

robe

0.68

enegger

0.68

 enthusi

0.66

rals

0.66

Activations Density 0.047%

words related to heightened states or intensification

terms related to heightened states of awareness or concern

No Comments

No Known Activations