INDEX

Explanations

terms related to large quantities or importance

oai_token-act-pair · gpt-3.5-turbo

terms related to significant amounts or collections of items

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

agent

-0.67

ivation

-0.65

angering

-0.64

 livest

-0.63

 effected

-0.62

TPPStreamerBot

-0.61

activ

-0.59

 interval

-0.59

activated

-0.59

active

-0.59

POSITIVE LOGITS

pload

0.70

gie

0.69

ãĤĮ

0.69

bilt

0.68

Â¯Â¯Â¯Â¯

0.67

above

0.66

enance

0.66

esome

0.65

sonian

0.64

dylib

0.64

Activations Density 0.067%

terms related to large quantities or importance

terms related to significant amounts or collections of items

No Comments

No Known Activations