INDEX

Explanations

words related to sharp objects or actions

oai_token-act-pair · gpt-3.5-turbo

occurrences of the distinct linguistic feature "th."

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

IFrame

Link

Negative Logits

ãĤŃ

-0.84

-0.79

ITED

-0.77

Tokens

-0.71

assetsadobe

-0.68

VICE

-0.67

HAEL

-0.67

FAULT

-0.67

ãĤ¹ãĥĪ

-0.65

 Katz

-0.65

POSITIVE LOGITS

ieving

1.11

istle

1.04

umb

1.01

umping

1.00

urst

0.97

orns

0.97

ttp

0.97

umbnails

0.95

ouse

0.94

warts

0.93

Activations Density 0.007%

words related to sharp objects or actions

occurrences of the distinct linguistic feature "th."

No Known Activations