INDEX

Explanations

words related to tracing, detecting, or uncovering something hidden or deep-rooted

oai_token-act-pair · gpt-3.5-turbo

references to the concept of tracing or tracking evidence

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

icum

-0.82

itch

-0.74

ivals

-0.67

haw

-0.67

hospital

-0.67

yss

-0.65

sky

-0.64

 spons

-0.62

demand

-0.62

skill

-0.62

POSITIVE LOGITS

 tracing

1.03

 traces

1.02

 trace

0.96

 traced

0.92

 ancestry

0.79

ogene

0.77

 footprints

0.77

ãĤ§

0.75

 bracelet

0.74

ability

0.74

Activations Density 0.047%

words related to tracing, detecting, or uncovering something hidden or deep-rooted

references to the concept of tracing or tracking evidence

No Comments

No Known Activations