INDEX

Explanations

terms related to time duration

oai_token-act-pair · gpt-3.5-turbo

phrases related to short-lived events or transient conditions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

ADRA

-0.82

chwitz

-0.69

EMENT

-0.66

iance

-0.66

 Democr

-0.64

olate

-0.64

ONSORED

-0.63

yss

-0.63

 Wasteland

-0.63

åĨ

-0.63

POSITIVE LOGITS

term

1.16

sighted

1.16

lived

1.09

comings

1.04

circ

0.92

duration

0.90

handed

0.88

period

0.88

 term

0.88

notice

0.83

Activations Density 0.050%

terms related to time duration

phrases related to short-lived events or transient conditions

No Comments

No Known Activations