INDEX

Explanations

information related to various topics

oai_token-act-pair · gpt-3.5-turbo

references to information in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

gg

-0.76

alone

-0.74

jug

-0.72

awar

-0.69

atos

-0.69

 Parables

-0.65

kus

-0.64

irth

-0.63

mount

-0.61

ensions

-0.61

POSITIVE LOGITS

 overload

0.91

 glean

0.86

 retrieval

0.83

 dissemination

0.81

ãĥĨ

0.79

 information

0.77

afety

0.77

ãĤ±

0.76

 theoret

0.76

frame

0.73

Activations Density 0.035%

information related to various topics

references to information in various contexts

No Comments

No Known Activations